1 / 19

A Scalable Execution Control Method for Context- dependent Services

A Scalable Execution Control Method for Context- dependent Services. Wataru Uchida, Hiroyuki Kasai, Shoji Kurakake Network Laboratories, NTT DoCoMo, Inc. Jun. 28, 2006. Outline. Background and motivation Proposal of service execution control method Simulation results

jared
Télécharger la présentation

A Scalable Execution Control Method for Context- dependent Services

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. A Scalable Execution Control Method for Context- dependent Services Wataru Uchida, Hiroyuki Kasai, Shoji Kurakake Network Laboratories,NTT DoCoMo, Inc. Jun. 28, 2006

  2. Outline • Background and motivation • Proposal of service execution control method • Simulation results • Conclusions and future works

  3. Background • Cellular networks are expected to provide context-dependent services • assist user's real world activities • continuously monitor context and are executed when the context satisfies pre-defined condition.

  4. Context-dependent services Child surveillance Push-based restaurant recommendation context: location of child context:・location of user・availability of tables I arrivedat the school! Frenchrestaurant"la mère"menu child's terminalwith GPS user mother'sterminal user terminal's display automatically notify mother of her child's arrival to school/station/private school. automatically recommends nearby restaurants which have vacant tables. Other examples: 24hours healthcare service, Friend-finder service,...

  5. Problem and objective • Need tremendous number of operations for execution controls • We have to • continuously acquire and collect many kinds of context • determine execution for a large number of services. • Example • Restaurant-recommendation: continuously locate user, measure number of vacant tables, collect them and determine to recommend or not • Execution control operations with low frequency doesn't always work well (risk of missing execution timing). Objective: reduce cost of execution control while preserving the service quality

  6. ② Context collection : execution control operations Service execution control ③ Determination of service execution Server A Server B Server C Executioncondition Executioncondition Executioncondition execution ・・・ Network ① Context acquisition Context acquisition terminals(ex. cell phones with GPS device, non-contact type IC cards,...) User

  7. Determination of execution • Calculate expected utility (EU) for execution(a1) and non-execution(a2), and chose one with higher EU • Utility: effect of execution/non-execution for the user • EU for a1 and a2: : state of context Expected utility (EU) :utility of action EU of non-execution ( ) execution time t EU of execution ( )

  8. Reduce execution control operations when probability of satisfying execution condition is low EU EU of non-execution ( ) EU of execution ( ) high frequency (large risk) low frequency (small risk of missing chance) principle of our methodReduction of execution control operations • Probability of satisfying execution condition (=risk of missing chance of execution) varies with time. t

  9. :probability distribution of EU estimated estimated Low probability High probability Interchange probability estimation • Predict context values • Utility in future can be estimated using predicted values • Compare estimated with EU of non-execution ( ) EU now t EU of execution ( )

  10. Collecting context with large effect • Interchange probability depends on the values of each context. • Each context's effect on the probability is not equal. • Context with large effect is collected more frequently. • Each context's effect can be calculated using conditional probabilities.

  11. Utility estimation • use Bayesian network • can handle probability distributions of context. a1: recommenda2: don't recommend Action (A) Distance(D) Option(O) Utility (U) Acceptsthe user(B) customernumber(C) Utility table

  12. Each terminal send values when the value enters "alert region" (estimation is incorrect and execution time approaches ) Collect context with high effect frequently when probability of satisfying execution conditionis high. (Future work) System architecture Server A Server B Server C ... Invoke execution Register execution condition Server-side controller Context 1 acquisition terminal Context 2 acquisition terminal ...

  13. Simulation setup • Metrics: number of collections, service quality (explained in following slides) • Assumed service: restaurant recommendation • Compared with: method which Periodically performs Execution Control operations (PEC) Context: distance from restaurant, availability of tables 3km Random-walk User 3km ・・・ Max speed: 100m/min Restaurant Num. vacant tables increased or decreased at every minute

  14. Service quality (1/2) • We measured execution ratio: (num. of timings service is executed) / (num. of timings execution condition is satisfied) • Service quality is high when the ratio is high. t Execution ratio = 4 / 8 = 50% :Timings execution condition is satisfied :Timings services are executed

  15. Service quality (2/2) • Also measured deviation from ideal decisions: • Sum of times when the decision is different from that of the ideal case (execution control with the highest frequency) • Service quality is high when the value is small. Time when the decision is different Timings in the ideal case t t Timings the method detected

  16. Reduce 90% of the total cost Result 1/2: Execution ratio Cost: high Number of collections Service quality: high

  17. Result 2/2:deviation from ideal decisions Cost: high Number of collections Service quality: high

  18. Conclusions • Scalable execution control method for context-dependent services • Methodology: gather context when the service execution condition is about to be satisfied • Simulation results: execution control operations are reduced while preserving service quality [Future works] • Development of execution control using alert region • Service quality loss-less (e.g. execution guaranteed) method

  19. Thank you! email: uchidaw@nttdocomo.co.jp

More Related