Summary of introduction to localization algorithms and algorithms

First, the electronic fence positioning algorithm:

or decided not to do the positioning algorithm, the reasons are as follows:

1. In the literature [1], a linear algorithm was used to solve the TDOA problem (four observation points above), and the form of the AOA algorithm was also included in the literature [2]. The intersection of multiple lines is the position of the point to be measured. If you consider whether a single point is inside the fence, the previous assumption is that the algorithm for locating a single point is highly complex, but these two articles illustrate that the theory is linear and simple. According to the following expansion of the problem into linear problem solving based on four or more observations, 1 combined with new application scenarios and actual data to obtain the results of the algorithm application test, like the literature [3] in the vehicle network in the application of this positioning algorithm, but the measured data I now very difficult to get. 2 will be a single point to multi-point, the most obvious feature of the Internet of Things is the massive access, if you can simultaneously determine whether multipoint is inside the fence, then this algorithm and the Internet of things have a stronger connection. I did not check the literature, but I could try, but the next reason for not doing it was more convincing. (It is very simple to deduce the scope and the constraints of each dimension difference within the fence, and the way to look up the table should be a particularly easy one, but the fault tolerance should be problematic.) ）

2. The positioning algorithm is not strongly related to the previous two modules in the whole thesis, and the first two algorithms, especially the task offload algorithm, are not very deep. So, instead of taking the time to add new parts, it's better to study the task offload problem a little bit further. Through the mobile cloud computing architecture, algorithms and applications, the task offload algorithm plays an important role in mobile cloud computing, and research should focus on this issue. In addition, the number of server virtual machines available in the book, The choice of cloud path for offload, is also a point that can be extended. It is more appropriate to add to the larger paper than to locate the algorithm.

Based on the above two points, the positioning algorithm first put aside.

[1] Chan Y T, Ho K C. A simple and efficient estimator for hyperbolic location[j]. IEEE transactions on Signal Processing, 1994, 42 (8): 1905-1915.

[2] Cong L, Zhuang W. Hybrid Tdoa/aoa Mobile User location for wideband CDMA Cellular Systems[j]. IEEE transactions on Wireless Communications, 2002, 1 (3): 439-447.

[3] Zhang K, Yang D, Miao F, et al. VEHICLE LOCALIZATION and TRACKING USING acous-tic sensors BASED on time difference of Arrival Method[j].

Ii. Summary of the introduction of algorithms

The introduction of the algorithm in the beginning of number 2nd, before looking at a lot of people's experience is "must be hands-on algorithm implementation" and "must do hands-on lessons after the problem." But I think this advice is not suitable for me, because I now particularly want to learn is the analysis algorithm and design algorithm method, to deepen their research depth. But if you want to find a job for algorithmic engineers, you should do it again. ok~ I'll do it again with a question algorithm.

So far, one of the most useful ideas is divide-conquer. Three methods of substituting, recursive tree and main method are used to analyze the algorithm running time. Another example of large circuit layout shows that the algorithm design process can be reversed by designing a run-time expression.

Read the book + See the MIT Open Class at NetEase. The Open class is really great, but it doesn't seem to cover everything in the book.

Before the design algorithm, feel oneself with really is "wild path", before completely by "algorithm introduction = Find work necessary" this kind of experience to mislead. Can learn the introduction of the algorithm feel like digging a treasure, especially good, is to do the groundwork for algorithm engineers ~ first put their own research point pad up, learning algorithm analysis and algorithm design, and then work to complement the algorithm to achieve. Learn something, come on.

Summary of introduction to localization algorithms and algorithms