Category: News
MATLAB Coder generated MEX file throws an error that “Index exceeds matrix dimensions” although the code runs in MATLAB without error
I’m trying to use MATLAB Coder to generate C/C++ code for my MATLAB code. However, when I try to execute the MEX file generated by MATLAB Coder, it throws an error that for a specific variable that the index exceeds matrix dimensions:
>> myCode_mex
Index exceeds matrix dimensions. The array ‘A’ is empty and therefore has no valid indices.
The error stack points to the following MATLAB source code where I perform array extension:
n = length(A);
if n < buffSize
A(n+1) = data;
The same code however runs in MATLAB without any issues. Why does this happen?I’m trying to use MATLAB Coder to generate C/C++ code for my MATLAB code. However, when I try to execute the MEX file generated by MATLAB Coder, it throws an error that for a specific variable that the index exceeds matrix dimensions:
>> myCode_mex
Index exceeds matrix dimensions. The array ‘A’ is empty and therefore has no valid indices.
The error stack points to the following MATLAB source code where I perform array extension:
n = length(A);
if n < buffSize
A(n+1) = data;
The same code however runs in MATLAB without any issues. Why does this happen? I’m trying to use MATLAB Coder to generate C/C++ code for my MATLAB code. However, when I try to execute the MEX file generated by MATLAB Coder, it throws an error that for a specific variable that the index exceeds matrix dimensions:
>> myCode_mex
Index exceeds matrix dimensions. The array ‘A’ is empty and therefore has no valid indices.
The error stack points to the following MATLAB source code where I perform array extension:
n = length(A);
if n < buffSize
A(n+1) = data;
The same code however runs in MATLAB without any issues. Why does this happen? array, index, exceeds, matrix, matlab, coder, extension, ml, codegen MATLAB Answers — New Questions
Confidence interval in fit
By default, the fit function in matlab returns 95 % confidence interval bounds. I want it to return error as defined by 1 sigma confidence, which i think corresponds to 68.27% confidence bound. How do i get this?By default, the fit function in matlab returns 95 % confidence interval bounds. I want it to return error as defined by 1 sigma confidence, which i think corresponds to 68.27% confidence bound. How do i get this? By default, the fit function in matlab returns 95 % confidence interval bounds. I want it to return error as defined by 1 sigma confidence, which i think corresponds to 68.27% confidence bound. How do i get this? curve fitting, nonlinear, error MATLAB Answers — New Questions
Method lookup in multiple inheritance
I defined four classes: a, b, c, and d. Class a defines a method f; class b inherits from a and overrides f; class c inherits from a; class d inherits from both b and c.
Why does calling f(d) print ‘b’? What is MATLAB’s method lookup mechanism?
classdef a
methods
function f(obj)
‘a’
end
end
end
classdef b < a
methods
function f(obj)
‘b’
end
end
end
classdef c < a
methods
end
end
classdef d < a
methods
end
endI defined four classes: a, b, c, and d. Class a defines a method f; class b inherits from a and overrides f; class c inherits from a; class d inherits from both b and c.
Why does calling f(d) print ‘b’? What is MATLAB’s method lookup mechanism?
classdef a
methods
function f(obj)
‘a’
end
end
end
classdef b < a
methods
function f(obj)
‘b’
end
end
end
classdef c < a
methods
end
end
classdef d < a
methods
end
end I defined four classes: a, b, c, and d. Class a defines a method f; class b inherits from a and overrides f; class c inherits from a; class d inherits from both b and c.
Why does calling f(d) print ‘b’? What is MATLAB’s method lookup mechanism?
classdef a
methods
function f(obj)
‘a’
end
end
end
classdef b < a
methods
function f(obj)
‘b’
end
end
end
classdef c < a
methods
end
end
classdef d < a
methods
end
end multiple inheritance MATLAB Answers — New Questions
The Sad State of Microsoft Graph and Other APIs
MVPs Express Annoyance and Frustration at the Sad State of the Microsoft Graph
An appeal by several MVPs in the Microsoft Graph feedback forum highlights the fragmented and incoherent state of APIs used across the Microsoft 365 ecosystem. The original promise that the Microsoft Graph would deliver a consistent method to access data distributed across many different data stores has never transpired, and developers and administrators for tenants, CSPs, and ISVs are frustrated by the seeming inability of Microsoft to bring order to the chaos created by Redmond.
Among the complaints made are:
Missing or inconsistent coverage of workloads by Graph APIs. Exchange Online has an Admin API, but that’s a sticking plaster to get past the need to remove EWS. SharePoint Online introduced a Graph API to manage tenant settings in 2022 and hasn’t done much since. At least SharePoint Online has an API to create sites (albeit limited): Exchange Online is limited to PowerShell when it comes to creating mailboxes. Many of the Teams administrative policies are based on the old Skype for Business framework and don’t have Graph APIs.
Widespread use of Undocumented APIs: Behind the scenes of many workloads are features inaccessible through PowerShell or a Microsoft Graph API. It’s understandable that engineering groups need time to complete the work to create a fully supported API, but it would also be nice if the major administrative portals used documented APIs whenever possible. For example, no interfaces are available for the critical Microsoft XDR solution.
Graph APIs remaining in beta for extended periods: Tenants understand that APIs need to meet performance and quality goals before they attain production status. Many tenants don’t like using beta APIs because they also understand that Microsoft can remove or drastically change functionality while APIs are in beta. It’s therefore unacceptable when APIs remain in beta for years. The AuditLogQuery API is an example. Microsoft has struggled to deliver a stable service for the synchronous audit search feature in PowerShell. They don’t seem to be able to get the asynchronous Graph API to run well either, even though it seems to be the one used by the Purview audit search solution. Tenants would like clarity about plans for moving beta APIs to production to help them plan their use of these APIs.
Lack of Resource for the Microsoft Graph PowerShell SDK: Many automated processes developed in tenants use PowerShell. The Microsoft Graph PowerShell SDK is an incredibly important tool. Each new version gets over 700,000 downloads. Despite the evidence of obvious customer interest and usage, instead of devoting sufficient engineering resources to maintain the Microsoft Graph PowerShell SDK properly so that reported bugs are fixed in a reasonable timeframe, the SDK is supported by a very small team. The result is poor quality and a large bug count.
It would also be nice for Microsoft to fix the ongoing problem of assembly clashes that affects major Microsoft 365 PowerShell modules. For example, I can load Exchange Online (V3.9.2) and then the Microsoft Graph PowerShell SDK (V2.35.1). but I can’t load the SDK followed by Exchange. How frustrating do you think this is to the average developer?
It is unacceptable for Microsoft to speak out of one side of its mouth by endorsing the Graph APIs as the route for external developers to interact with Microsoft 365 while at the same time declining to support the development and enhancement of the APIs.
Copilot Sucks Resources
Many long-time observers see only one cause to blame. Microsoft’s focus on artificial intelligence and integrating Copilot into literally every piece of software they produce has stolen engineering budget and resources from other tasks. Developers working on projects not aligned with Copilot tell of budget and headcount cuts with small teams given new responsibilities without extra resources. Over coffees and beers, I’ve heard about teams being cut in half while also being asked to take on new projects. Allied to cuts in writers, the net effect is declining quality in both code and documentation.
Customers aren’t stupid. We see the lack of responsiveness to software bugs. We see the lack of consistency across APIs and the holes that appear in APIs. We look at the documentation and wonder how anyone could create such horrible examples. We cope with poor support that isn’t getting any better despite the fact that understanding Office 365 and cloud-based services are no longer like searching for a hidden secret. And we see contacts in Microsoft who used to work on Graph-related topics being moved to other areas with no replacement.
The Real State of Microsoft 365
Microsoft obviously believes all is well within Microsoft 365. The service meets its financially backed SLA every quarter, so everything must be OK. The quarterly results for the Microsoft Cloud are applauded by Wall Street analysts because of high profitability and growing numbers. It’s nice for CFO Amy Hood to see her much-loved Average Revenue Per User (ARPU) increase as the result of efforts to upsell licenses and add-ons, and those numbers will get a nice uptick in July 2026 when Microsoft increases monthly license fees. Not that the increases will come close to offsetting the investment Microsoft has made to build out datacenter capacity for AI services.
The profitability of the Microsoft Cloud brings no joy for customers, especially those who eschew the notion of spending $360/year/user on Microsoft 365 Copilot (or $1,188 for the new Microsoft 365 E7 license). The 3.3% or so of the Microsoft 365 installed base who pay for Copilot must be happy because of the rapid pace of development to deliver new features. The rest of the user base is paying for that development without benefiting from the enormous software engineering effort.
We know Microsoft can do great things when the company sets a target. The ability to take on and solve technical challenges is engrained into Microsoft history. All that is being asked here is to help the Microsoft 365 community by bringing stability to and predictability to the Microsoft Graph APIs and associated tools.
The Way Forward
It wouldn’t take much to solve the issues reported here. The world’s largest software engineering company has the resources, talent, and funding to make rapid improvements in the Microsoft Graph and associated tools.
Microsoft management should make it a priority to restore consistency and robustness to the Microsoft Graph, to ensure that sufficient resources are in place to improve quality and close the gaps in current coverage, and to make the Microsoft Graph APIs deliver on the vision laid out years ago. When that happens, customers can get on with the task of building automated processes to fill in the gaps left by Microsoft.
Help Microsoft make the right decision by expressing your opinions and voting for the requests expressed in the Microsoft Graph feedback forum. Your opinion counts!
Support the work of the Office 365 for IT Pros team by subscribing to the Office 365 for IT Pros eBook. Your support pays for the time we need to track, analyze, and document the changing world of Microsoft 365 and Office 365. Only humans contribute to our work!
Microsoft at NVIDIA GTC: New solutions for Microsoft Foundry, Azure AI infrastructure and Physical AI
Microsoft combines accelerated computing with cloud scale engineering to bring advanced AI capabilities to our customers. For years, we’ve worked with NVIDIA to integrate hardware, software and infrastructure to power many of today’s most important AI breakthroughs.
What’s new at NVIDIA GTC
- Expanded Microsoft Foundry capabilities to build, deploy and operate production-ready AI agents on NVIDIA accelerators and open NVIDIA Nemotron models
- New Azure AI infrastructure optimized for inference-heavy, reasoning-based workloads, including the first hyperscale cloud to power on next-generation NVIDIA Vera Rubin NVL72 systems
- Deeper integration across Microsoft Foundry, Microsoft Fabric and NVIDIA Omniverse libraries and open frameworks to support Physical AI systems from simulation to real‑world operations
From Frontier models to production-ready agents
At the foundation of this system is Microsoft Foundry: serving as the operating system for building, deploying and operating AI at enterprise scale. Foundry builds on Azure to bring together models, tools, data and observability into a single system designed for production agents. Today we’re expanding those capabilities across Foundry Agent Service and NVIDIA Nemotron models.
The next-generation Foundry Agent Service and Observability in Foundry Control Plane are now generally available, enabling organizations to build and operate AI agents at production scale. Foundry Agent Service allows teams to quickly develop agents that reason, plan and act across tools, data and workflows. Once created, Foundry Control Plane provides the developer end-to-end visibility into agent behavior, unlocking both developer productivity as well as enterprise trust. Companies such as Corvus Energy are already using Foundry to replace manual inspection workflows with agent-driven operational intelligence across their global fleet.
We are further simplifying the path from prototype to production with the availability of Voice Live API integration with Foundry Agent Service, in public preview, which enables developers to build voice-first, multimodal, real-time agentic experiences. This pairs with the general availability of a refreshed Microsoft Foundry portal and expanded integrations for Palo Alto Networks’ Prisma AIRS and Zenity, delivering deeper builder experiences and runtime security across the entire agent lifecycle.
NVIDIA Nemotron models are also now available through Microsoft Foundry, joining the widest selection of models on any cloud, including the latest reasoning, frontier and open models. This bolsters our recent partnership announcement bringing Fireworks AI to Microsoft Foundry, enabling customers to fine-tune open-weight models like NVIDIA Nemotron into low-latency assets that can be distributed to the edge.
Scaling AI infrastructure for the world’s most demanding workloads
Inference AI workloads are reshaping cost, performance and system design requirements. To operationalize agentic AI at scale, customers need purpose-built infrastructure for inference‑heavy, reasoning‑based workloads that can be deployed and operated consistently across global and regulated environments.
Microsoft’s AI infrastructure approach is engineered to seamlessly bring next-generation NVIDIA systems into Azure datacenters that are designed for power, cooling networking and rapid generational upgrades. This allows our customers to move with speed and agility and stay at the leading edge from generation to generation.
In less than a year, we’ve deployed hundreds of thousands of liquid-cooled Grace Blackwell GPUs across our global datacenter footprint, and now we are excited to be the first hyperscale cloud to power on NVIDIA’s newest Vera Rubin NVL72 in our labs. Over the next few months, Vera Rubin NVL72 will be rolled out into our modern, liquid-cooled Azure datacenters.
Microsoft’s infrastructure innovation with NVIDIA also extends to sovereign and regulated environments to give customers control of both where AI runs and how it evolves over time. Recently, we announced Foundry Local support for modern infrastructure and large AI models, and today we now have initial support for NVIDIA Vera Rubin platform on Azure Local, extending accelerated AI capabilities to customer-controlled environments. This approach allows organizations to plan for next-generation AI workloads, including reasoning-based and agentic systems, while maintaining Azure-consistent operations, governance and security through our unified software layer with Azure Arc and Foundry Local.
Bringing AI into the physical world
As AI moves beyond digital experiences, Microsoft and NVIDIA are collaborating to support the next wave of Physical AI. At GTC, this work centers on NVIDIA Physical AI Data Factory Blueprint, with Microsoft Foundry as the platform for hosting and operating Physical AI systems on Azure at cloud scale.
By integrating this blueprint with Azure services as part of a Physical AI Toolchain, Microsoft enables developers to build, train and operate physical AI and robotics workflows that connect physical assets, simulation and cloud training environments into repeatable, enterprise-grade pipelines. To support, we are introducing a public Azure Physical AI Toolchain GitHub repository integrated with the Nvidia Physical AI Data Factory and with core Azure services.
To further the impact of AI in real‑world, physical environments, today Microsoft and NVIDIA are deepening the integration between Microsoft Fabric and NVIDIA Omniverse libraries, connecting live operational data with physically accurate digital twins and simulation. This allows organizations to see what’s happening across their physical systems, understand it in real time and use AI to decide what to do next. In practice, customers in manufacturing and operations and beyond are using this approach to move beyond dashboards and alerts to coordinated, AI‑driven action across machines, facilities and workflows.
From innovation to impact
Microsoft is delivering reliable, production‑scale AI by bringing together its global AI infrastructure, platforms and real‑world systems with the latest innovation from NVIDIA. For customers, this means the ability to operate intelligence continuously, running inference-heavy, reasoning-based and physical AI workloads with the performance, security and governance required for real businesses and regulated industries.
Whether powering always-on agents, scaling next-generation AI infrastructure or deploying intelligent systems in factories, energy facilities and sovereign environments, Microsoft and Nvidia are helping customers move faster from insight to action.
Yina Arenas leads product strategy and execution for Microsoft Foundry, overseeing the end–to–end AI product portfolio, infrastructure, developer experiences and foundation model integration across OpenAI, Anthropic, Mistral, DeepSeek and others. She delivers an enterprise ready, production grade AI platform trusted by global customers for secure, reliable and scalable AI.
The post Microsoft at NVIDIA GTC: New solutions for Microsoft Foundry, Azure AI infrastructure and Physical AI appeared first on The Official Microsoft Blog.
Microsoft combines accelerated computing with cloud scale engineering to bring advanced AI capabilities to our customers. For years, we’ve worked with NVIDIA to integrate hardware, software and infrastructure to power many of today’s most important AI breakthroughs. What’s new at NVIDIA GTC Expanded Microsoft Foundry capabilities to build, deploy and operate production-ready AI agents on…
The post Microsoft at NVIDIA GTC: New solutions for Microsoft Foundry, Azure AI infrastructure and Physical AI appeared first on The Official Microsoft Blog.Read More
Some figures not saving as vector graphics (.svg)
Hi,
I ran into a strange problem. I’m trying to save figures in vector format to process in Adobe Illustrator. I make numerous of these plots using one script and in many cases the output to vector graphics using the .svg format works fine. Some figures, made by the same script and apart from the specific datapoints indistinguishable from other figures, do not export as proper vector graphics. In those cases I can open the output .svg file in Illustrator but the image is basically flat; i.e. I can not isolate individual elements for editing.
I have attached two figures extracted at the exact point in the script where the data is added that causes it to not save properly. I plot datapoints on a trial-trial basis, adding data from one trial, then the data from the next trial etc. Until a particular trial I can save the figure as a proper vector graphic .svg file (attached as Fig1Vector.fig). When data from the next trial is added, the output .svg file only contains a flattened image (attached as Fig2NotVector.fig).
I’ve been looking for differences between the two figures but I have exhausted my ideas on what could be at cause. Do any of you have any idea what could be going wrong here or any suggestions to do a successful vector image export?
Thanks in advance.
Best,
PaulHi,
I ran into a strange problem. I’m trying to save figures in vector format to process in Adobe Illustrator. I make numerous of these plots using one script and in many cases the output to vector graphics using the .svg format works fine. Some figures, made by the same script and apart from the specific datapoints indistinguishable from other figures, do not export as proper vector graphics. In those cases I can open the output .svg file in Illustrator but the image is basically flat; i.e. I can not isolate individual elements for editing.
I have attached two figures extracted at the exact point in the script where the data is added that causes it to not save properly. I plot datapoints on a trial-trial basis, adding data from one trial, then the data from the next trial etc. Until a particular trial I can save the figure as a proper vector graphic .svg file (attached as Fig1Vector.fig). When data from the next trial is added, the output .svg file only contains a flattened image (attached as Fig2NotVector.fig).
I’ve been looking for differences between the two figures but I have exhausted my ideas on what could be at cause. Do any of you have any idea what could be going wrong here or any suggestions to do a successful vector image export?
Thanks in advance.
Best,
Paul Hi,
I ran into a strange problem. I’m trying to save figures in vector format to process in Adobe Illustrator. I make numerous of these plots using one script and in many cases the output to vector graphics using the .svg format works fine. Some figures, made by the same script and apart from the specific datapoints indistinguishable from other figures, do not export as proper vector graphics. In those cases I can open the output .svg file in Illustrator but the image is basically flat; i.e. I can not isolate individual elements for editing.
I have attached two figures extracted at the exact point in the script where the data is added that causes it to not save properly. I plot datapoints on a trial-trial basis, adding data from one trial, then the data from the next trial etc. Until a particular trial I can save the figure as a proper vector graphic .svg file (attached as Fig1Vector.fig). When data from the next trial is added, the output .svg file only contains a flattened image (attached as Fig2NotVector.fig).
I’ve been looking for differences between the two figures but I have exhausted my ideas on what could be at cause. Do any of you have any idea what could be going wrong here or any suggestions to do a successful vector image export?
Thanks in advance.
Best,
Paul figure, export, saveas, plot, save, vector graphics, svg, fig MATLAB Answers — New Questions
Section folding not present
Using R2024B, i got the option to use the live editor and some things changed without asking.
Most importantly, the + buttons to fold a section have dissappeared, while in preferences/editor/code folding the option sections is ON.
Another change is that the titles of the scripts are in an additional column.
FInally, beneath the folder section, there is a new window with the ‘script’ , i.e. a shortcut to jump to a section.
How can I get the section folding again?
Cheers, PaulUsing R2024B, i got the option to use the live editor and some things changed without asking.
Most importantly, the + buttons to fold a section have dissappeared, while in preferences/editor/code folding the option sections is ON.
Another change is that the titles of the scripts are in an additional column.
FInally, beneath the folder section, there is a new window with the ‘script’ , i.e. a shortcut to jump to a section.
How can I get the section folding again?
Cheers, Paul Using R2024B, i got the option to use the live editor and some things changed without asking.
Most importantly, the + buttons to fold a section have dissappeared, while in preferences/editor/code folding the option sections is ON.
Another change is that the titles of the scripts are in an additional column.
FInally, beneath the folder section, there is a new window with the ‘script’ , i.e. a shortcut to jump to a section.
How can I get the section folding again?
Cheers, Paul section folding MATLAB Answers — New Questions
what’s wrong with matlab dsolve ?
I am using this matlab version
‘24.2.0.2806996 (R2024b) Update
Example:
——————————————-
syms u(t) a
sl = dsolve( diff(u,t) == sqrt( u^2+a^2) )
sl =
-a*1i
a*1i
exp(C1 + t)/2 – (a^2*exp(- C1 – t))/2
——————————————
Apart from the constant solution,
the correct solution is
u(t) = a*sinh( t+ const.)
which is different than the matlab solution unless a^2 = 1
What’s wrong with dsolve ?I am using this matlab version
‘24.2.0.2806996 (R2024b) Update
Example:
——————————————-
syms u(t) a
sl = dsolve( diff(u,t) == sqrt( u^2+a^2) )
sl =
-a*1i
a*1i
exp(C1 + t)/2 – (a^2*exp(- C1 – t))/2
——————————————
Apart from the constant solution,
the correct solution is
u(t) = a*sinh( t+ const.)
which is different than the matlab solution unless a^2 = 1
What’s wrong with dsolve ? I am using this matlab version
‘24.2.0.2806996 (R2024b) Update
Example:
——————————————-
syms u(t) a
sl = dsolve( diff(u,t) == sqrt( u^2+a^2) )
sl =
-a*1i
a*1i
exp(C1 + t)/2 – (a^2*exp(- C1 – t))/2
——————————————
Apart from the constant solution,
the correct solution is
u(t) = a*sinh( t+ const.)
which is different than the matlab solution unless a^2 = 1
What’s wrong with dsolve ? dsolve MATLAB Answers — New Questions
Is the New Agent 365 Worth $99?
Checking Out Microsoft 365 E7: E5, Copilot, the Entra Suite, and Agent 365
On March 9, Microsoft unveiled what it calls the “first Frontier Suite” — Microsoft 365 E7. At $99 per user per month, E7 bundles Microsoft 365 E5, Microsoft 365 Copilot, the Entra Suite, and a brand-new product called Agent 365 into a single SKU. General availability is set for May 1, 2026. This announcement naturally raises the question of whether US$99/month is a worthwhile investment or not.
What’s in the Box
Microsoft 365 E7 wraps together components that many E5 tenants already use (or aspire to use). The bundle includes:
- Microsoft 365 E5 (which continues to include Defender, Purview, Intune, Teams, Exchange, SharePoint, and the rest)
- Microsoft 365 Copilot
- Microsoft Entra Suite (Entra ID P2, Entra Internet Access, Entra Private Access, and the Entra Suite governance features)
- Agent 365 — a new add-on, priced separately at $15/user/month, that promises governance, security, and observability for AI agents
Microsoft’s messaging positions E7 as the suite for a “human-led, agent-operated enterprise,” powered by something called “Work IQ” — a shared intelligence layer that feeds signals across applications, Copilot, and agents. Jared Spataro describes Agent 365 as “one place to observe, secure, and govern every agent across the organization.”
Agent 365 Deserves Scrutiny
Agent 365 is the genuinely new piece here. It extends aspects of Entra ID, Purview, and Defender to cover AI agents in addition to human users. That means agents built in Copilot Studio, Azure AI Foundry, or through the new Agent 365 framework can be assigned identities, have their actions audited, be subjected to compliance policies, and monitored for risky behavior — much the same way you manage human users today.
The practical implications are significant. If your organization has started building or deploying agents (and Microsoft is betting heavily that you will), you’ll need a governance layer. Agent 365 is Microsoft’s pitch that the governance layer should be theirs, not a third party’s. For example, Agent 365 can automatically expire agents that haven’t been used, and it can flag agents that don’t have a listed owner— basic governance functionality we have today for Groups, Teams, SharePoint sites, and other objects, but which haven’t previously been available for agents.
Doing the License Math
If you’re already licensing E5 and Copilot separately, E7 could represent modest savings, especially once you factor in the Entra Suite. The combination of Copilot and a Microsoft 365 E5 license lists at US$90, so for the extra US$9 you also get Agent 365 (US$15 standalone) and Entra Suite (US$12 standalone). Whether this makes sense or not depends on two things.
First, will you actually use the additional features? Copilot adoption has been slow, with Microsoft’s own figures at their FY26 Q2 results putting the number at around 3.3% of the Microsoft 365 installed base (more than 450 million paid seats). If you’re not willing to pay US$30/month for standalone Copilot, it’s questionable whether $39/month for the whole suite of E7 features is more compelling. Microsoft’s recent announcements of “wave 3” Copilot features may help settle this question if the new features prove useful.
Second, you’ll probably still have to pay to run your agents. Microsoft’s documentation says that Copilot 365 license holders don’t have to pay to “use agents in Microsoft 365 services or apps,” create “classic answers [or] generative answers,” or answer prompts that are grounded by Microsoft Graph in the tenant. For every other agent, though, you’ll be paying (and possibly paying twice if you’re using Foundry, since its usage by developers will be billed as Azure consumption and not through Copilot Studio.)
Microsoft has been running various promotional discount programs (example, example, example) for Copilot throughout 2025 and into 2026, so it will be interesting to see if they continue this trend.
Partner Discounts Are on the Table
Microsoft is also sweetening the deal for partners by offering some new co-sell tools. More importantly, E7 and Agent 365 will be eligible workloads for CSP incentives, which Microsoft often uses as a way to drive deployment of new workloads or upsell features.
What IT Pros Should Do Now
- Audit your current licensing. Compare what you’re paying today for E5 + Copilot + Entra components against the $99 E7 price point. The break-even analysis will vary by tenant.
- Assess your agent readiness. If you aren’t building agents yet, E7 may be premature. If you are, Agent 365’s governance capabilities could save you from a future security headache.
- Watch the Work IQ story carefully. Microsoft is building a shared intelligence layer across all its products. That’s either a competitive moat or a lock-in play, depending on your perspective.
- Don’t rush. GA is May 1. Use the intervening weeks to model costs and evaluate whether Agent 365’s capabilities address risks you actually face.
The E7 announcement is a clear signal that Microsoft sees agents as the next growth vector for Microsoft 365. Whether that future arrives on schedule — and whether tenants are ready for the price tag — remains to be seen.
Support the work of the Office 365 for IT Pros team by subscribing to the Office 365 for IT Pros eBook. Your support pays for the time we need to track, analyze, and document the changing world of Microsoft 365 and Office 365. Only humans contribute to our work!
Why is there not a button for FIK and License file?
It’s impossible to find these 2 pieces of info. Please create a button pointing to them next to the license number.It’s impossible to find these 2 pieces of info. Please create a button pointing to them next to the license number. It’s impossible to find these 2 pieces of info. Please create a button pointing to them next to the license number. fik MATLAB Answers — New Questions
Modeling load-depend friction in bearings in Simscape
Hi everyone,
i’m trying to model a five-point double-toggle lever with bearings with load-dependent friction, based on the normal-force (Coulomb, which is constant at any velocity).
I’ve tried modeling the bearing with the Multibody Interface and Rotational Friction, but i can only add a constant friction torque. When i calculate the torque using the measured normal force of the bearing, I get an algebraic loop. Is it possible to apply a load dependent Torque (Tr = Fr*d/2 = Fn*µ*d/2) and a friction force (Fr = µ*Fn) without creating algebraic loops?
Note on the image below: The force Rij is calculated as R = µ*N and acts opposite to the velocity. If R is transformed to the center, this results in atorque (T = r*R), acting opposite to the velocity.
Best greetings!Hi everyone,
i’m trying to model a five-point double-toggle lever with bearings with load-dependent friction, based on the normal-force (Coulomb, which is constant at any velocity).
I’ve tried modeling the bearing with the Multibody Interface and Rotational Friction, but i can only add a constant friction torque. When i calculate the torque using the measured normal force of the bearing, I get an algebraic loop. Is it possible to apply a load dependent Torque (Tr = Fr*d/2 = Fn*µ*d/2) and a friction force (Fr = µ*Fn) without creating algebraic loops?
Note on the image below: The force Rij is calculated as R = µ*N and acts opposite to the velocity. If R is transformed to the center, this results in atorque (T = r*R), acting opposite to the velocity.
Best greetings! Hi everyone,
i’m trying to model a five-point double-toggle lever with bearings with load-dependent friction, based on the normal-force (Coulomb, which is constant at any velocity).
I’ve tried modeling the bearing with the Multibody Interface and Rotational Friction, but i can only add a constant friction torque. When i calculate the torque using the measured normal force of the bearing, I get an algebraic loop. Is it possible to apply a load dependent Torque (Tr = Fr*d/2 = Fn*µ*d/2) and a friction force (Fr = µ*Fn) without creating algebraic loops?
Note on the image below: The force Rij is calculated as R = µ*N and acts opposite to the velocity. If R is transformed to the center, this results in atorque (T = r*R), acting opposite to the velocity.
Best greetings! simulink, friction, bearing, simscape, loop MATLAB Answers — New Questions
Custom storage class for Argument Simulink parameters identifiable by TLC script
Hi,
My objective is to define a custom storage class specifically intended for "Argument" Simulink parameters. The primary requirement is to programmatically identify which model parameters are using this custom storage class from within a TLC script executed during code generation.
I initially attempted to create the custom storage class using an Embedded Coder Dictionary. However, at code generation time, I have been unable to definitively determine which model parameters are using my custom storage class. The only observable indicator appears to be the field ModelParameters.Parameter[i].IsSpecifiedAsNewStorageClass, which is set to 1. I would appreciate confirmation on whether this is the intended mechanism or if additional fields should be accessible.
I subsequently explored the Custom Storage Class Designer, which showed promise. Upon inspection, the custom storage class appeared clearly within the TLC script context under the CustomStorageClasses record. However, as far as I know, it is not possible to define a custom storage class for "Argument" Simulink parameters this way.
Are there any aspects of either methodology that I may have overlooked or configured incorrectly?
Is there an alternative approach or recommended candidate method that would better satisfy my requirements?
Best regardsHi,
My objective is to define a custom storage class specifically intended for "Argument" Simulink parameters. The primary requirement is to programmatically identify which model parameters are using this custom storage class from within a TLC script executed during code generation.
I initially attempted to create the custom storage class using an Embedded Coder Dictionary. However, at code generation time, I have been unable to definitively determine which model parameters are using my custom storage class. The only observable indicator appears to be the field ModelParameters.Parameter[i].IsSpecifiedAsNewStorageClass, which is set to 1. I would appreciate confirmation on whether this is the intended mechanism or if additional fields should be accessible.
I subsequently explored the Custom Storage Class Designer, which showed promise. Upon inspection, the custom storage class appeared clearly within the TLC script context under the CustomStorageClasses record. However, as far as I know, it is not possible to define a custom storage class for "Argument" Simulink parameters this way.
Are there any aspects of either methodology that I may have overlooked or configured incorrectly?
Is there an alternative approach or recommended candidate method that would better satisfy my requirements?
Best regards Hi,
My objective is to define a custom storage class specifically intended for "Argument" Simulink parameters. The primary requirement is to programmatically identify which model parameters are using this custom storage class from within a TLC script executed during code generation.
I initially attempted to create the custom storage class using an Embedded Coder Dictionary. However, at code generation time, I have been unable to definitively determine which model parameters are using my custom storage class. The only observable indicator appears to be the field ModelParameters.Parameter[i].IsSpecifiedAsNewStorageClass, which is set to 1. I would appreciate confirmation on whether this is the intended mechanism or if additional fields should be accessible.
I subsequently explored the Custom Storage Class Designer, which showed promise. Upon inspection, the custom storage class appeared clearly within the TLC script context under the CustomStorageClasses record. However, as far as I know, it is not possible to define a custom storage class for "Argument" Simulink parameters this way.
Are there any aspects of either methodology that I may have overlooked or configured incorrectly?
Is there an alternative approach or recommended candidate method that would better satisfy my requirements?
Best regards code generation, simulink parameters, storage class MATLAB Answers — New Questions
DLL fails to load – matlabruntimeforpython_abi3
Hi all,
I’ve been struggling with this one for a few days now. I’m looking for proof of concept program that I can package a matlab function and then import it to my python executable. So far everything runs great in a terminal, but after using PyInstaller to generate the executable the I cannot get it to run. I’ve installed the MATLAB 2025b runtime and did install the package that was generated using the Python Package Compiler in MATLAB.
I get the following output in my command window when I attempt to launch matlab_test.exe:
My MATLAB function:
function a = matlab2python(b)
a = sqrt(b);
end
My python code:
import os
import sys
import time
import matlabsdk as sdk_python_test
import matlab
if __name__ == "__main__":
pkg = sdk_python_test.initialize()
result = pkg.matlab2python(9.0)
print(result)
pkg.terminate()Hi all,
I’ve been struggling with this one for a few days now. I’m looking for proof of concept program that I can package a matlab function and then import it to my python executable. So far everything runs great in a terminal, but after using PyInstaller to generate the executable the I cannot get it to run. I’ve installed the MATLAB 2025b runtime and did install the package that was generated using the Python Package Compiler in MATLAB.
I get the following output in my command window when I attempt to launch matlab_test.exe:
My MATLAB function:
function a = matlab2python(b)
a = sqrt(b);
end
My python code:
import os
import sys
import time
import matlabsdk as sdk_python_test
import matlab
if __name__ == "__main__":
pkg = sdk_python_test.initialize()
result = pkg.matlab2python(9.0)
print(result)
pkg.terminate() Hi all,
I’ve been struggling with this one for a few days now. I’m looking for proof of concept program that I can package a matlab function and then import it to my python executable. So far everything runs great in a terminal, but after using PyInstaller to generate the executable the I cannot get it to run. I’ve installed the MATLAB 2025b runtime and did install the package that was generated using the Python Package Compiler in MATLAB.
I get the following output in my command window when I attempt to launch matlab_test.exe:
My MATLAB function:
function a = matlab2python(b)
a = sqrt(b);
end
My python code:
import os
import sys
import time
import matlabsdk as sdk_python_test
import matlab
if __name__ == "__main__":
pkg = sdk_python_test.initialize()
result = pkg.matlab2python(9.0)
print(result)
pkg.terminate() pyinstaller, python312 MATLAB Answers — New Questions
Teams Meetings to Block Third-Party Recording Bots
Stop Third-Party Recording Bots Generating Transcription and AI Summaries
Message center notification MC1251206 (13 March 2026, Microsoft 365 roadmap item 558107) is very interesting. It announces the introduction of third-party recording bot detection and blocking in Teams meetings. According to Microsoft, a profusion of third-party “meeting assistant bots” that join meetings and use the audio stream to transcribe and summarize meetings are in increasing use.
Microsoft plans to deploy the feature to targeted release tenants starting in mid-May 2026. General availability should follow in early June 2026 to complete by mid-June. GCC will also get the feature in the same timeframe. Bot detection becomes part of the core Teams infrastructure and will be enabled for all tenants.
The Preference for Native Transcription
Naturally, Microsoft would like you to use the transcription feature built into Teams to create transcripts of Teams meetings. Their point is that use of native transcription is clearly flagged through warning notices shown to meeting participants. Meeting transcripts are stored in OneDrive for Business in the MP4 files generated for Teams recordings (with or without video) and available for eDiscovery. If you want summarization (meeting notes and action items), Copilot for Teams can do that, and its meeting notes are available for compliance purposes too. Such are the benefits of using integrated features within an ecosystem like Microsoft 365.
By comparison, Microsoft says that some third-party “bots may access meetings without the knowledge or consent of the meeting organizer or the hosting tenant, which can create data security, privacy, and compliance risks.” They have a point. I have joined many Teams meetings, including some in the Microsoft tenant, where an odd participant shows up that’s been added by a human participant.
Detecting the Presence of Third-Party Recording Bots
The new capability allows Teams to check participants as they attempt to join meetings to detect the presence of bots. If meeting organizers are happy for people to use personal bots to capture details of what goes on and store that information outside of your tenant, they can admit the bot from the meeting lobby to join the meeting (Figure 1) and do its stuff. I don’t think this is a sensible idea.
Microsoft admits that bot detection is not perfect and might not pick up every third-party recording bot that’s currently available. However, over time they reckon that customer reports and their own research will allow the detection algorithm to become more accurate and dependable. And besides, if meeting organizers notice strange participants, they can always eject those participants to protect the information discussed in the meeting.
Administrative Control
Microsoft says that a new meeting policy will be available in the Teams admin center to allow administrators to control how bot detection works. I believe this to mean that Microsoft plans to add a new control to the Teams meeting policy that will have values such as disable bot detection and require approval to admit bots. The policy is not yet available and V7.6 of the Teams PowerShell module doesn’t show a setting for the Set-CsTeamsMeetingPolicy cmdlet either. Microsoft says that they plan to provide more granular controls in the future.
A Question of Balance
Cynics will say that blocking third-party recording bots is a blatant attempt by Microsoft to protect their ability to sell Microsoft 365 Copilot licenses to the 96% of the 450 million-plus installed base that haven’t yet bought Microsoft 365 Copilot. Microsoft disagrees and points to the obvious compliance issues that arise when potentially confidential data is taken outside the tenant to be processed to generate meeting notes and summaries.
On balance, I think Microsoft is right to introduce third-party recording bot detection and to make the feature part of the base Teams software. If tenants want to allow the use of third-party recording bots, they can update their Teams meeting policies to disable the block.
Learn about managing Teams and the rest of the Microsoft 365 ecosystem by subscribing to the Office 365 for IT Pros eBook. Use our experience to understand what’s important and how best to protect your tenant.
Flexibility in Satellite Toolbox
Is there any way to both add and visualize customized gimbals and/or sensors to the satellite toolbox?
My particular scenario is looking at putting a camera on a platform but limiting the gimbal angles and field-of-view non-symmetrically.
The FOV is different in along and cross track directions (not conical). While other toolboxes have "rectangular" FOVs, I don’t see the capability in the satellite toolbox. Functions could be added to filter, but they won’t visualize properly.
My gimbals are pitch and roll rather than azimuth and elevation. I can compute the corresponding angles, but wondering if there’s a way to change how gimbals are controlled.
My gimbals don’t have complete freedom to move throughout the entire hemisphere. Again, results can be filtered but the visualization doesn’t correspond.
Is there currently (or possibly in the future) a path to customizing gimbals or sensors (with visualization) in this toolbox? Maybe an API?Is there any way to both add and visualize customized gimbals and/or sensors to the satellite toolbox?
My particular scenario is looking at putting a camera on a platform but limiting the gimbal angles and field-of-view non-symmetrically.
The FOV is different in along and cross track directions (not conical). While other toolboxes have "rectangular" FOVs, I don’t see the capability in the satellite toolbox. Functions could be added to filter, but they won’t visualize properly.
My gimbals are pitch and roll rather than azimuth and elevation. I can compute the corresponding angles, but wondering if there’s a way to change how gimbals are controlled.
My gimbals don’t have complete freedom to move throughout the entire hemisphere. Again, results can be filtered but the visualization doesn’t correspond.
Is there currently (or possibly in the future) a path to customizing gimbals or sensors (with visualization) in this toolbox? Maybe an API? Is there any way to both add and visualize customized gimbals and/or sensors to the satellite toolbox?
My particular scenario is looking at putting a camera on a platform but limiting the gimbal angles and field-of-view non-symmetrically.
The FOV is different in along and cross track directions (not conical). While other toolboxes have "rectangular" FOVs, I don’t see the capability in the satellite toolbox. Functions could be added to filter, but they won’t visualize properly.
My gimbals are pitch and roll rather than azimuth and elevation. I can compute the corresponding angles, but wondering if there’s a way to change how gimbals are controlled.
My gimbals don’t have complete freedom to move throughout the entire hemisphere. Again, results can be filtered but the visualization doesn’t correspond.
Is there currently (or possibly in the future) a path to customizing gimbals or sensors (with visualization) in this toolbox? Maybe an API? satellitetoolbox, gimbal, concialsensor MATLAB Answers — New Questions
Simulink simulation becomes very slow after adding FOC to a 6-phase PMSM drive with CHB inverter
I am simulating an asymmetrical six-phase PMSM drive in Simulink (R2025b).
The original model consisted of:
– 5-level cascaded H-bridge inverter
– phase-shifted PWM (PSPWM)
– asymmetrical 6-phase PMSM (30° phase shift)
In this configuration the simulation runs normally. After introducing field-oriented control (FOC) the simulation becomes very slow.
The FOC implementation includes:
– VSD transformation (abc-def → αβxy)
– Park transformation
– PI current controllers (id, iq, ix, iy)
– speed PI controller
– inverse transformations
– discrete current filters ( for i_αβ)
Main parameters:
– PWM carrier frequency: fcr = 5 kHz
– duty cycle update rate: 100 kHz
– solver: ode14x
– Simulink version: R2025b
The electrical model of the inverter and machine has not changed, only the control structure was added.
My question is: Is it normal that adding FOC significantly slows down the simulation in Simulink, and what are recommended practices to keep such motor drive simulations efficient (solver choice, controller sampling rates, or implementation of transformations)?I am simulating an asymmetrical six-phase PMSM drive in Simulink (R2025b).
The original model consisted of:
– 5-level cascaded H-bridge inverter
– phase-shifted PWM (PSPWM)
– asymmetrical 6-phase PMSM (30° phase shift)
In this configuration the simulation runs normally. After introducing field-oriented control (FOC) the simulation becomes very slow.
The FOC implementation includes:
– VSD transformation (abc-def → αβxy)
– Park transformation
– PI current controllers (id, iq, ix, iy)
– speed PI controller
– inverse transformations
– discrete current filters ( for i_αβ)
Main parameters:
– PWM carrier frequency: fcr = 5 kHz
– duty cycle update rate: 100 kHz
– solver: ode14x
– Simulink version: R2025b
The electrical model of the inverter and machine has not changed, only the control structure was added.
My question is: Is it normal that adding FOC significantly slows down the simulation in Simulink, and what are recommended practices to keep such motor drive simulations efficient (solver choice, controller sampling rates, or implementation of transformations)? I am simulating an asymmetrical six-phase PMSM drive in Simulink (R2025b).
The original model consisted of:
– 5-level cascaded H-bridge inverter
– phase-shifted PWM (PSPWM)
– asymmetrical 6-phase PMSM (30° phase shift)
In this configuration the simulation runs normally. After introducing field-oriented control (FOC) the simulation becomes very slow.
The FOC implementation includes:
– VSD transformation (abc-def → αβxy)
– Park transformation
– PI current controllers (id, iq, ix, iy)
– speed PI controller
– inverse transformations
– discrete current filters ( for i_αβ)
Main parameters:
– PWM carrier frequency: fcr = 5 kHz
– duty cycle update rate: 100 kHz
– solver: ode14x
– Simulink version: R2025b
The electrical model of the inverter and machine has not changed, only the control structure was added.
My question is: Is it normal that adding FOC significantly slows down the simulation in Simulink, and what are recommended practices to keep such motor drive simulations efficient (solver choice, controller sampling rates, or implementation of transformations)? simulink, motor-control, pmsm, foc, power_electronics_control MATLAB Answers — New Questions
How to place code generated from subsystem into RAM on Ti C2000?
I am trying to plase parts of the code generated from one of the subsystems into RAM to speed up execution. The help provided in Control Data and Function Placement in Memory by Inserting Pragmas is not sufficient.
Is there a better explanation on how to add #pragma CODE_SECTION(myCriticalTask, "ramfuncs") to the generated function?
Do I need to edit linker command file?
Do I need to add code to copy function from flash to RAM on power up?
Thanks.I am trying to plase parts of the code generated from one of the subsystems into RAM to speed up execution. The help provided in Control Data and Function Placement in Memory by Inserting Pragmas is not sufficient.
Is there a better explanation on how to add #pragma CODE_SECTION(myCriticalTask, "ramfuncs") to the generated function?
Do I need to edit linker command file?
Do I need to add code to copy function from flash to RAM on power up?
Thanks. I am trying to plase parts of the code generated from one of the subsystems into RAM to speed up execution. The help provided in Control Data and Function Placement in Memory by Inserting Pragmas is not sufficient.
Is there a better explanation on how to add #pragma CODE_SECTION(myCriticalTask, "ramfuncs") to the generated function?
Do I need to edit linker command file?
Do I need to add code to copy function from flash to RAM on power up?
Thanks. c2000, matlab, simulink, embedded coder MATLAB Answers — New Questions
Clarification on Sampling Rate vs. Control Bandwidth in Average Current Mode Control for Buck Converter
I am implementing Average Current Mode Control for a Buck converter. According to control theory, the voltage loop (outer loop) is typically designed to be 5–10 times slower than the current loop (inner loop) to ensure stability and proper decoupling.
However, in my current simulation setup, the ADC for both voltage and current is triggered at the same time. Does this imply that the execution frequency (sampling rate) of both control loops must be identical?
How can I implement a slower voltage loop while maintaining the same ADC sampling frequency for both signals in Simulink/Embedded C? Any insights on whether this ‘simultaneous sampling’ contradicts the principle of having a slower outer loop would be greatly appreciatedI am implementing Average Current Mode Control for a Buck converter. According to control theory, the voltage loop (outer loop) is typically designed to be 5–10 times slower than the current loop (inner loop) to ensure stability and proper decoupling.
However, in my current simulation setup, the ADC for both voltage and current is triggered at the same time. Does this imply that the execution frequency (sampling rate) of both control loops must be identical?
How can I implement a slower voltage loop while maintaining the same ADC sampling frequency for both signals in Simulink/Embedded C? Any insights on whether this ‘simultaneous sampling’ contradicts the principle of having a slower outer loop would be greatly appreciated I am implementing Average Current Mode Control for a Buck converter. According to control theory, the voltage loop (outer loop) is typically designed to be 5–10 times slower than the current loop (inner loop) to ensure stability and proper decoupling.
However, in my current simulation setup, the ADC for both voltage and current is triggered at the same time. Does this imply that the execution frequency (sampling rate) of both control loops must be identical?
How can I implement a slower voltage loop while maintaining the same ADC sampling frequency for both signals in Simulink/Embedded C? Any insights on whether this ‘simultaneous sampling’ contradicts the principle of having a slower outer loop would be greatly appreciated average current mode, cascaded control MATLAB Answers — New Questions
Amplitude Modulation using simulink
I am currently doing amplitude modulation using three very high frequency sine waves which I getting perfectly after adding all the three due to already set parameters in the simulation parameter box.
The carrier frequency is also very high, up to 3000 Hz, but the scope is displaying is correctly. But after the message signal and carrier are passing through the product, I am not getting amplitude modulated signal. Instead, I am getting a completely zero straight line. My sample time in the Product is 0.0001.
I am pretty sure it has to do something with the product parameters or simulation parameters but I so not sure exactly what it is?I am currently doing amplitude modulation using three very high frequency sine waves which I getting perfectly after adding all the three due to already set parameters in the simulation parameter box.
The carrier frequency is also very high, up to 3000 Hz, but the scope is displaying is correctly. But after the message signal and carrier are passing through the product, I am not getting amplitude modulated signal. Instead, I am getting a completely zero straight line. My sample time in the Product is 0.0001.
I am pretty sure it has to do something with the product parameters or simulation parameters but I so not sure exactly what it is? I am currently doing amplitude modulation using three very high frequency sine waves which I getting perfectly after adding all the three due to already set parameters in the simulation parameter box.
The carrier frequency is also very high, up to 3000 Hz, but the scope is displaying is correctly. But after the message signal and carrier are passing through the product, I am not getting amplitude modulated signal. Instead, I am getting a completely zero straight line. My sample time in the Product is 0.0001.
I am pretty sure it has to do something with the product parameters or simulation parameters but I so not sure exactly what it is? amplitude, sine MATLAB Answers — New Questions
how to add noise with vairience 0.002 in the code with mean 0
a = -1;
b = 1;
g = a + (b-a).*rand(1,250);
x = zeros(1,250);
k = zeros(1,250);
v = 0.01*randn(1,250);
for n = 1:250
k(n) = g(n) + 0.5*g(n)*g(n)*g(n);
end
for t = 6:250
x(t) = 0.8*(x(t-1) – 0.3*x(t-2) – 0.105*x(t-3) + 0.059*x(t-4)) …
+ 0.6*(x(t-2) – 0.3*x(t-3) – 0.105*x(t-4) + 0.059*x(t-5)) …
+ 0.4*(k(t-1) – 0.3*k(t-2) – 0.105*k(t-3) + 0.059*k(t-4)) …
– 0.2*(k(t-2) – 0.3*k(t-3) – 0.105*k(t-4) + 0.059*k(t-5)) …
+ v(t);
end
now i have to generate noise signal with variance 0.002 and mean 0,how to write codea = -1;
b = 1;
g = a + (b-a).*rand(1,250);
x = zeros(1,250);
k = zeros(1,250);
v = 0.01*randn(1,250);
for n = 1:250
k(n) = g(n) + 0.5*g(n)*g(n)*g(n);
end
for t = 6:250
x(t) = 0.8*(x(t-1) – 0.3*x(t-2) – 0.105*x(t-3) + 0.059*x(t-4)) …
+ 0.6*(x(t-2) – 0.3*x(t-3) – 0.105*x(t-4) + 0.059*x(t-5)) …
+ 0.4*(k(t-1) – 0.3*k(t-2) – 0.105*k(t-3) + 0.059*k(t-4)) …
– 0.2*(k(t-2) – 0.3*k(t-3) – 0.105*k(t-4) + 0.059*k(t-5)) …
+ v(t);
end
now i have to generate noise signal with variance 0.002 and mean 0,how to write code a = -1;
b = 1;
g = a + (b-a).*rand(1,250);
x = zeros(1,250);
k = zeros(1,250);
v = 0.01*randn(1,250);
for n = 1:250
k(n) = g(n) + 0.5*g(n)*g(n)*g(n);
end
for t = 6:250
x(t) = 0.8*(x(t-1) – 0.3*x(t-2) – 0.105*x(t-3) + 0.059*x(t-4)) …
+ 0.6*(x(t-2) – 0.3*x(t-3) – 0.105*x(t-4) + 0.059*x(t-5)) …
+ 0.4*(k(t-1) – 0.3*k(t-2) – 0.105*k(t-3) + 0.059*k(t-4)) …
– 0.2*(k(t-2) – 0.3*k(t-3) – 0.105*k(t-4) + 0.059*k(t-5)) …
+ v(t);
end
now i have to generate noise signal with variance 0.002 and mean 0,how to write code noise MATLAB Answers — New Questions
