Maximum bandwidth used in function of obliged Tx threshold 

Context in source publication

Context 1

... to the higher computational time, one of the main cited drawbacks of following an unstructured approach for SF resources assignment is the high amount of signaling needed in order to broadcast the assignments to the terminals, because the division of the frame in carriers can completely change on an SF_T basis. But it must also be considered that the satellite links are quite asymmetric, where the capacity in the forward direction (from ground networks to satellite terminals) is usually much higher than in the return link direction. Whether or not the amount of signaling required by the proposed heuristic is acceptable or excessive will depend on the characteristics of the considered communication network, but it is an important point to take into account in the design trade-offs of the satellite network return and forward links. Hereafter are compared the results obtained for each file when the obliged transmissions threshold is set to the other values considered in Table 1, which are the lowest possible value, i.e. SF_T and to a value between the minimum and the previously reported values for PLR=0 (see last column of Table 3). Looking at Figure 16, there does not seems to be a dependency between the resulting maximum bandwidth used and the value set for the obliged transmissions threshold. The same can be said of efficiency, according to Figure 17. The PLR due to transmitter overload always decreases as the obliged transmission threshold increases, as shown in Figure 18. It has been found, by tracing the software, that the transmitter overloads are due to the expiration of streams of packets that must be transmitted continuously, without any gap time in the transmission, and which arrive at a rate exceeding the maximum transmission rate, e.g. arrival of 60 ms duration bursts each 20 ms. As explained in Juan et al (2010), several instances of the proposed packing heuristic, based on the BFDH heuristic could be launched in parallel using different alpha values uniformly distributed in the range (0.10, 0.025). Each heuristic, during the solution-construction process, at the time a new burst and packing level must be selected from the horizontal space remaining sorted list of levels, would apply the following probability distribution for the random variable X = “level k-th is selected at the current step”, where k=1,2,...,s, with s being the current size of the list: P(X = k) = alpha · (1 – alpha) k-1 + epsilon, for any k = 1,2,...,s where epsilon = 1 - sum(alpha · (1 – alpha) k-1 , k=1..s) is a term that assigns a positive probability to every possible step, because of this it is a quasi-geometric distribution. Notice that if the size s of levels list is large enough, the term epsilon is close to zero, therefore the alpha parameter can be interpreted as the probability of selecting the step with the best match to the BFDH heuristic matching criteria. Instances of the heuristic executed with a low alpha value (e.g. 0.05) consider a large number of levels from the list when selecting the next step, while instances running with a relatively high alpha-value (e.g. 0.35) use a more reduced list of potentially eligible levels. The randomized versions, each one with its corresponding alpha parameter configured, would select the next step by generating a random number and would get the position of the next step to be selected using a geometric distribution. The solution using the deterministic version of the heuristic would be generated also in parallel with the randomized versions. It would be used as an upper bound limit of what is considered a good solution in terms of bandwidth utilization. As described before, our approach could make use of parallel execution of the heuristic to generate a set of random feasible solutions. Each solution can be computed in a few milliseconds using a PC. It has been shown the design and performance of a new heuristic to solve the resource-assignment problem on telecommunications networks for time critical communications, assuming an unstructured approach to the problem. It is based on the BFDH heuristic for 2D packing problems. This is a novel and innovative approach to the RRM in telecommunication networks in two dimensions. First, because unstructured approaches are frequently discarded in telecommunications literature due to computational cost, in favor of more structured approaches, easier to implement and characterize. But this paper shows that the unstructured approach to the problem is feasible for considered representative traffic profiles, without the need of extraordinary computing resources. On the other hand, it is well-known that innovations usually happen by merging two or more separate topics. This paper algorithm is quite innovative in its approach of solving a telecommunications networks problem from an operations research problem perspective, adapting an existing heuristic to the field of strip packing problems (BFDH) to the telecommunications networks resources management, by establishing an analogy between orders of strips of materials and the need to transmit RF bursts packed on a frame of time and spectrum. Açar, G. (2001). “End-to-end resource management in geostationary satellite networks”. Ph.D. dissertation, Dept. Elect. Electron. Eng., Imperial Coll., London, U.K. Fairhurst, G. (April 2001). “Broadband Satellite Overview”. http :// www . abdn . ac . uk / erg / research / future - net / digital - video / bband - sat . html . Retrieved on March 2012. ETSI (09/2005) EN 301 790 v1.4.1. “Digital Video Broadcasting (DVB); Interaction channel for satellite distribution systems”. ETSI (03/2005) TR 101 790 v1.2.1 “Digital Video Broadcasting (DVB); Interaction channel for Satellite Distribution Systems; Guidelines for the use of EN 301 790 ”. Roberts, L. G., Lawrence G. (April 1975). "ALOHA Packet System With and Without Slots and Capture". Computer Communications Review 5 (2): 28–42 Shannon, Claude E. (July/October 1948). “A Mathematical Theory of Communication”. Bell System Technical Journal 27 (3): 379-423. Indra, UPC GRCM. (10/2010). “ANTARES Communication Standard Design Justification. ...