Distributed Computation

Loading...

來自 Princeton University 的課程

Networks Illustrated: Principles without Calculus

40 個評分

Princeton University

Networks Illustrated: Principles without Calculus

40 個評分

What makes WiFi faster at home than at a coffee shop? How does Google order its search results from the trillions of webpages on the Internet? Why does Verizon charge $15 for every GB of data we use? Is it really true that we are connected in six social steps or less? These are just a few of the many intriguing questions we can ask about the social and technical networks that form integral parts of our daily lives. This course is about exploring the answers, using a language that anyone can understand. We will focus on fundamental principles like “sharing is hard”, “crowds are wise”, and “network of networks” that have guided the design and sustainability of today’s networks, and summarize the theories behind everything from the social connections we make on platforms like Facebook to the technology upon which these websites run. Unlike other networking courses, the mathematics included here are no more complicated than adding and multiplying numbers. While mathematical details are necessary to fully specify the algorithms and systems we investigate, they are not required to understand the main ideas. We use illustrations, analogies, and anecdotes about networks as pedagogical tools in lieu of detailed equations. Please note that per Princeton University policy, no certificates, credentials or reports are awarded in connection with this course.

從本節課中

Power Control in Cellular Networks

How is it possible that we can all communicate effectively without disrupting each other's calls, messages, or Internet usage? In this lesson, we will take a look at some of the methods that have been developed for letting us "share" the air over which our phones communicate.

與講師見面

Christopher Brinton
Lecturer
Electrical Engineering
Mung Chiang
Professor
Electrical Engineering

So we just got done illustrating Qualcom's famous algorithm called

Distributive Power Control, which solves the near-far problem.

And it has a lot of characteristics of it that are very unique, and very nice and

make it a very sophisticated algorithm that's easy to run.

The first one, is that, it is completely, distributed.

As we said the first word is distributed. So what exactly do we mean, by

distributed? Well the idea is that, the tower, really

doesn't have to do, much work at all. I mean, the computation is all happening,

in each of these phones right here. All the tower has to do is measure all

the SIR's, and then send them back to each of the cellphones.

And each of the devices on the device side are going to actually update their

own power levels. That's the first thing.

The first point of exterior computation is that there's no centralized mechanism.

There's no big supercomputer in here that is doing all the computation.

There's none of that. So, all the computation happens on this

end at the device site. Now, the second part of distributed

computation. Is that, none of these devices have to

have any operation of what the other devices are doing.

All this device needs to know is what it's measured SIR is [SOUND] because the

effect that each of these devices are imposing on this device [SOUND] is given

to it in the form of that, this measured SIR.

because it effects this. But it only needs to know its own

measured SIR. It doesn't need to know anybody else's

measured SIR. Thus needs the only conditions one need

is channel conditions and anybody else's current transfer powers.

Nothing. All it needs to know is are its own

parameters. So when we say distributed computation we

mean two things. First is that there's no centralized

computation. It lies on device side to do all of the

math and programming. And the second is that no device needs no

knowledge of what the devices are doing. So, DPC is the solution to the near-far

problem and has many other characteristics which are desirable as

well. First is that it's highly scalable and

that's result of it being distributed. The way to see that is to note that each

of these devices are doing the computation.

So, every time we put another device in the cell, it's not going to add any other

strain on some centralized mechanism because we don't have something like

that. So, as we will see later on in the course

achieving scalability in networks is a very Important thing.

And making things distributed is a way of achieving that scalability.

Because every device that comes into the network gets to do its own work.

And doesn't have to have something in the server doing its work for it.

Another thing is that it is very inexpensive, in terms of computation.

It's just a simple one step algorithm. It just takes that measured SAR and the

desired SAR. And it takes the ratio, and it multiplies

it by the current transmit power. And that's all it needs to do over and

over again. And it really needs to be inexpensive

because it's done something like 1500 times each second.

So something happening that quickly really needs to be inexpensive.

And it needs to be able to be done very fast.

So that the levels can reconverge quickly.

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2018 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取