انجام یک پروژه با متلب و AdvantEDGE
چهار سال پیش منتشر شده
تعداد بازدید: 335
کد پروژه: 181003
شرح پروژه
لطفا بر اساس فایل های ضمیمه قیمت و زمان انجام پروژه را تعیین فرمایید.
In this work, M-MIMO at BS is assumed with NOMA schemes for 2 users in uploading scenario where the objective of optimization is to reduce the total latency including the local data processing. End-to-end system model with mathematical model is described with problem formulation to study delay in M-MIMO with NOMA system under MEC scheme. First, we defined the effect of uploading in the case of M-MIMO system, so the channel coefficient should be clarified. We studied the deadline for computation for near and far users and data rate for far user since it’s the most effected user as data rate perspective. We used energy consumption model that suit NOMA system. Then, to take power allocation coefficient for NOMA system in the consideration also in NOMA, the number of transmitted bits to MEC is different, so we compute it for near and far users. Finally, applied the optimization for total delay under different power values at far user under energy consumption constraint. Figure 1 illustrates the system model and the main parameters.
The problem formulation and the network model has been theatrically analyzed where a set, U of mobile users associated with a BS. The BS will use M-MIMO and NOMA to receive the messages sent by the users. To reduce the system complexity, suppose that NOMA successive interference cancellation (SIC) is only applied to a pair of users, . We will discuss how to select two users as a pair in NOMA SIC.
Assume the task arrival rate at user i is , and a task contains bits of data. User i’s task processing rate is . A proportion of tasks are offloaded to BS, and proportion of tasks are processed locally.
We consider a M/M/1 queuing system. For the tasks processed by user i locally, the average service delay (total time a task spends in the system including the time spent waiting and executing) does not depend on scheduling discipline and can be computed using Little’s law.
The Problem optimization and the distributed ADMM-VS formulated.
To make clearer, I am studying the impact of M-MIMO NOMA on the delay and energy consumption in the task offloading for edge computing. In order to obtain insight into the performance of the proposed M-MIMO NOMA offloading optimization scheme, in this section, I focus on the special case that two users offload their tasks to an edge server with M-MIMO NOMA transmission techniques. The optimization problem is formulated, in which the proportion of the tasks to be offloaded to the edge server as well as the transmit power used for offloading by each of the users are determined to minimize the total average service delay under the power consumption constraints.
For simulation
I want to build a simulation platform and doing the simulations based on Alg. 2 formulation I developed. He needs to have several baseline scenarios to compare,
1) Massive MIMO-NOMA
2) NOMA only
3) Massive MIMO only
4) OMA offloading
5) Full offloading
6) local computing only
The reference for baselines and simulation figures check the R3 attached,
In terms of wireless network parameters, check reference R1,R3 and R4 attached.
I need the performance figures:
Ave. Task Service Response Time vs. #of Task Arrival Rate per Second for different algorithms. similar to Figure 6 in R0
Ave. # of tasks offloaded by a node vs. differerent # of task arrival rate per second for different algorithms, similar to Figure 7 in R0
Ave. Energy Consumption per task vs. #of Task Arrival rate per second for different algorithms, similar to Figure 6 in R3
Check the references for the simulation setup, parameters, and figures.
He wants to use AdvantEDGE emulator because It provides a more realistic environment.
You can have one node as a BS, and two nodes as mobile devices, and then implement the algorithms within these nodes and set the channel conditions.
So you have to simulate three nodes, one BS with edge server, two mobiles. You can use AdvanEDGE emulator to build these three nodes, and their channel conditions between the three nodes. Then you can run the algorithms in each node. Or You can have three containers (emulate a BS and 2 mobiles). Each container can run its algorithm. You can implement the distributed algorithms in the container. You can try to run/port Matlab algorithms in AdvantEDGE containers. AdvanEDGE can be configured to emulate the channel conditions and communications.
So basically, we use AdvantEDGE to create three containers for 1 BS and 2 mobiles. Then run the algorithms in the containers (containers are like virtual machines). You can try whether you can run Matlab algorithms you have directly in the containers or need to port Matlab algorithms to containers. That is, the distributed ADMM algorithms run in three nodes and exchange messages over AdvantEdge channel to achieve the optimization, instead of pure Matlab simulation.
I will try to provide with everything that’s only keep coding for you. Please ask any questions.
Thankyou
این پروژه شامل 2 فایل مهم است، لطفا قبل از ارسال پیشنهاد حتما نسبت به بررسی این فایل اقدام فرمایید.
مهارت ها و تخصص های مورد نیاز
مهلت برای انجام
20روز
وضعیت مناقصه
بسته
درباره کارفرما
عضویت پنج سال پیش
