Wireless_NN1 Car_NN1 Sensors_NN2 Vulnerable_JJ to_II Hackers_NP1 Researchers_NN2 figure_VV0 out_RP how_RRQ to_TO hijack_VVI sensor_NN1 communications_NN2 ._. 
Hackers_NN2 could_VM "_" hijack_VV0 "_" the_AT wireless_NN1 pressure_NN1 sensors_NN2 built_VVN into_II many_DA2 cars_NN2 '_GE tires_VVZ ,_, researchers_NN2 have_VH0 found_VVN ._. 
Criminals_NN2 might_VM then_RT track_VVI a_AT1 vehicle_NN1 or_CC force_VV0 its_APPGE electronic_JJ control_NN1 system_NN1 to_II malfunction_NN1 ,_, the_AT University_NN1 of_IO South_NP1 Carolina_NP1 and_CC Rutgers_NP1 University_NN1 researchers_NN2 say_VV0 ._. 
The_AT team_NN1 ,_, which_DDQ successfully_RR hijacked_VVD two_MC popular_JJ tire-pressure-monitoring_JJ systems_NN2 (_( TPMS_NP1 )_) ,_, will_VM describe_VVI the_AT work_NN1 at_II the_AT USENIX_NN1 Security_NN1 conference_NN1 in_II Washington_NP1 ,_, DC_NP1 ,_, this_DD1 week_NNT1 ._. 
The_AT tire-sensor_JJ attack_NN1 poses_VVZ little_DA1 immediate_JJ risk_NN1 to_II drivers_NN2 ._. 
However_RR ,_, in_II recent_JJ months_NNT2 ,_, research_NN1 groups_NN2 have_VH0 identified_VVN other_JJ security_NN1 weaknesses_NN2 in_II vehicle_NN1 electronics_NN1 systems_NN2 ._. 
As_CSA automakers_NN2 add_VV0 more_RGR powerful_JJ computers_NN2 to_II cars_NN2 ,_, and_CC connect_VV0 those_DD2 computers_NN2 to_II critical_JJ components_NN2 ,_, in-car_JJ systems_NN2 will_VM need_VVI to_TO be_VBI secured_VVN against_II hackers_NN2 ,_, experts_NN2 warn_VV0 ._. 
A_ZZ1 TPMS_NP1 consists_VVZ of_IO sensors_NN2 inside_II a_AT1 car_NN1 's_GE tires_VVZ that_CST measure_VV0 pressure_NN1 ,_, and_CC a_AT1 central_JJ wireless_NN1 antenna_NN1 --_NN1 or_CC an_AT1 antenna_NN1 in_II each_DD1 wheel_NN1 in_II more_RGR expensive_JJ vehicles_NN2 ._. 
An_AT1 electric_JJ control_NN1 unit_NN1 (_( ECU_NN )_) picks_VVZ up_RP the_AT signal_NN1 ,_, and_CC a_AT1 warning_NN1 light_NN1 on_II the_AT automobile_NN1 's_GE dashboard_NN1 warns_VVZ a_AT1 driver_NN1 when_RRQ tire_VV0 pressure_NN1 has_VHZ dropped_VVN ._. 
As_II31 well_II32 as_II33 calculating_VVG pressure_NN1 changes_NN2 ,_, the_AT ECU_NN filters_VVZ out_RP noise_NN1 from_II sensors_NN2 in_II neighboring_VVG cars_NN2 ,_, and_CC compensates_VVZ for_IF pressure_NN1 changes_NN2 due_II21 to_II22 temperature_NN1 ._. 
The_AT TREAD_VV0 Act_NN1 ,_, which_DDQ Congress_NN1 passed_VVN in_II 2008_MC ,_, mandates_NN2 that_CST all_DB new_JJ vehicles_NN2 produced_VVN or_CC sold_VVN in_II the_AT United_NP1 States_NP1 after_II that_DD1 year_NNT1 are_VBR required_VVN to_TO have_VHI this_DD1 technology_NN1 ._. 
Using_VVG equipment_NN1 costing_VVG $1,500_NNU ,_, including_II a_AT1 programmable_JJ radio_NN1 transmitter_NN1 ,_, a_AT1 specialized_JJ circuit_NN1 board_NN1 ,_, and_CC free_JJ software_NN1 ,_, the_AT South_ND1 Carolina-Rutgers_NP1 team_NN1 could_VM pick_VVI up_RP a_AT1 car_NN1 's_GE tire_VV0 pressure_NN1 readings_NN2 ._. 
The_AT researchers_NN2 deciphered_VVD the_AT communication_NN1 protocol_NN1 by_II experimenting_VVG with_IW different_JJ parameters_NN2 of_IO the_AT radio_NN1 transmission_NN1 ._. 
The_AT systems_NN2 tested_VVN by_II the_AT South_ND1 Carolina-Rutgers_NP1 team_NN1 had_VHD very_RG little_DA1 security_NN1 in_II place_NN1 --_NN1 they_PPHS2 mainly_RR relied_VVN on_II the_AT fact_NN1 that_CST the_AT communications_NN2 protocol_NN1 is_VBZ not_XX widely_RR published_VVN ._. 
"_" In_II doing_VDG TPMS_NP1 this_DD1 way_NN1 ,_, &lsqb;_( automakers_NN2 &rsqb;_) have_VH0 left_VVN the_AT door_NN1 open_JJ to_II wireless_NN1 attackers_NN2 ,_, "_" says_VVZ Travis_NP1 Taylor_NP1 ,_, one_MC1 of_IO the_AT paper_NN1 's_GE authors_NN2 ._. 
The_AT team_NN1 could_VM eavesdrop_VVI on_II communications_NN2 and_CC ,_, in_II some_DD circumstances_NN2 ,_, alter_VV0 messages_NN2 in-transit_NN1 ._. 
That_DD1 let_VVD the_AT team_NN1 give_VVI false_JJ readings_NN2 to_II a_AT1 car_NN1 's_GE dashboard_NN1 ._. 
They_PPHS2 could_VM also_RR track_VVI a_AT1 vehicle_NN1 's_GE movements_NN2 using_VVG the_AT unique_JJ IDs_NN2 of_IO the_AT pressure_NN1 sensors_NN2 ,_, and_CC even_RR cause_VV0 a_AT1 car_NN1 's_GE ECU_NN to_TO fail_VVI completely_RR ._. 
"_" Normally_RR ,_, these_DD2 &lsqb;_( attacks_NN2 would_VM &rsqb;_) result_VV0 in_II small_JJ problems_NN2 ,_, "_" Taylor_NP1 says_VVZ ._. 
"_" But_CCB I_PPIS1 see_VV0 practical_JJ danger_NN1 and_CC damage_VV0 that_CST can_VM happen_VVI from_II TPMS_NP1 exploitation_NN1 ._. "_" 
Earlier_RRR this_DD1 year_NNT1 ,_, researchers_NN2 from_II the_AT University_NN1 of_IO Washington_NP1 and_CC the_AT University_NN1 of_IO California_NP1 ,_, San_NP1 Diego_NP1 showed_VVD that_CST they_PPHS2 could_VM take_VVI over_II the_AT control_NN1 systems_NN2 of_IO a_AT1 popular_JJ model_NN1 of_IO automobile_NN1 ,_, causing_VVG the_AT brakes_NN2 to_TO lock_VVI or_CC the_AT engine_NN1 to_TO cut_VVI out_RP ._. 
"_" The_AT security_NN1 and_CC privacy_NN1 problems_NN2 that_CST the_AT authors_NN2 identify_VV0 in_II TPMS_NP1 systems_NN2 are_VBR likely_JJ just_RR one_MC1 among_II many_DA2 that_CST will_VM challenge_VVI the_AT automotive_JJ industry_NN1 in_II the_AT years_NNT2 to_TO come_VVI ,_, "_" says_VVZ Stefan_NP1 Savage_NP1 ,_, a_AT1 UC_NP1 San_NP1 Diego_NP1 professor_NN1 of_IO computer_NN1 science_NN1 and_CC engineering_NN1 and_CC an_AT1 author_NN1 of_IO the_AT earlier_JJR report_NN1 ._. 
ECUs_NNU2 entered_JJ production_NN1 vehicles_NN2 in_II the_AT 1970s_MC2 ,_, following_VVG the_AT California_NP1 Clean_JJ Air_NN1 Act_NN1 and_CC a_AT1 surge_NN1 in_II gas_NN1 prices_NN2 ._. 
At_RR21 first_RR22 ,_, the_AT systems_NN2 were_VBDR just_RR used_VVN to_TO adjust_VVI the_AT fuel-oxygen_JJ mix_NN1 using_VVG data_NN from_II the_AT vehicle_NN1 's_GE exhaust_NN1 ._. 
But_CCB the_AT use_NN1 of_IO ECUs_NNU2 has_VHZ expanded_VVN since_II then_RT ,_, and_CC today_RT they_PPHS2 are_VBR used_VVN in_II every_AT1 aspect_NN1 of_IO monitoring_NN1 and_CC controlling_VVG automobiles_NN2 ._. 
ECUs_NNU2 are_VBR responsible_JJ for_IF a_AT1 feature_NN1 known_VVN as_II roll_NN1 stability_NN1 control_NN1 --_NN1 they_PPHS2 can_VM apply_VVI the_AT brakes_NN2 ,_, reduce_VV0 the_AT throttle_NN1 ,_, and_CC modulate_VV0 the_AT steering_NN1 to_TO keep_VVI a_AT1 car_NN1 from_II rolling_VVG over_RP ._. 
The_AT South_ND1 Carolina-Rutgers_NP1 team_NN1 stresses_VVZ that_CST it_PPH1 would_VM be_VBI difficult_JJ to_TO attack_VVI a_AT1 car_NN1 through_II the_AT wireless_NN1 tire_VV0 system_NN1 ._. 
One_MC1 hurdle_NN1 is_VBZ that_CST the_AT tire_VV0 sensors_NN2 communicate_VV0 infrequently_RR --_JJ about_RG once_RR every_AT1 60_MC to_II 90_MC seconds_NNT2 --_NN1 making_VVG it_PPH1 difficult_JJ to_TO manipulate_VVI the_AT system_NN1 ,_, especially_RR if_CS a_AT1 vehicle_NN1 is_VBZ moving_VVG ._. 
They_PPHS2 were_VBDR able_JK to_TO overcome_VVI these_DD2 hurdles_NN2 ,_, however_RR ,_, by_II shadowing_VVG a_AT1 target_NN1 vehicle_NN1 ,_, and_CC using_VVG directional_JJ antennas_NN2 ._. 
Even_RR21 so_RR22 ,_, UCSD_NP1 's_GE Savage_NP1 stresses_VVZ that_DD1 car-hacking_NN1 is_VBZ a_AT1 theoretical_JJ threat_NN1 for_IF the_AT moment_NN1 ._. 
"_" One_PN1 should_VM n't_XX overreact_VVI to_II this_DD1 kind_NN1 of_IO news_NN1 ,_, "_" says_VVZ Savage_NP1 ._. 
"_" It_PPH1 's_VBZ not_XX the_AT case_NN1 ,_, for_REX21 example_REX22 ,_, that_CST the_AT authors_NN2 have_VH0 identified_VVN a_AT1 means_NN by_II which_DDQ the_AT TPMS_NP1 channel_NN1 can_VM remotely_RR compromise_VVI safety-critical_JJ systems_NN2 ,_, nor_CC is_VBZ there_EX any_DD evidence_NN1 that_CST this_DD1 channel_NN1 is_VBZ being_VBG targeted_VVN ,_, or_CC even_RR that_CST there_EX is_VBZ a_AT1 clear_JJ threat_NN1 of_IO it_PPH1 being_VBG a_AT1 likely_JJ target_NN1 in_II the_AT near-term_JJ ._. "_" 
Savage_NP1 says_VVZ his_APPGE group_NN1 has_VHZ entered_VVN discussions_NN2 with_IW the_AT "_" appropriate_JJ stakeholders_NN2 "_" regarding_II the_AT exploits_NN2 they_PPHS2 discovered_VVD ._. 
The_AT South_ND1 Carolina-Rutgers_NP1 team_NN1 has_VHZ had_VHN no_AT luck_NN1 so_RG far_RR in_II contacting_VVG carmakers_NN2 ._. 
The_AT Alliance_NN1 of_IO Automobile_NN1 Manufacturers_NN2 will_VM take_VVI the_AT appropriate_JJ steps_NN2 to_TO secure_VVI its_APPGE vehicles_NN2 ,_, says_VVZ spokesman_NN1 Wade_NP1 Newton_NP1 ._. 
"_" While_CS this_DD1 concern_NN1 is_VBZ n't_XX unique_JJ to_II autos_NN2 ,_, we_PPIS2 continue_VV0 looking_VVG at_II this_DD1 issue_NN1 --_NN1 before_CS it_PPH1 becomes_VVZ a_AT1 problem_NN1 ,_, "_" he_PPHS1 says_VVZ ._. 
Sensors_NN2 for_IF Tracking_JJ Home_NN1 Water_NN1 Use_NN1 Sensors_NN2 track_VV0 devices_NN2 '_GE electricity_NN1 ,_, water_NN1 ,_, and_CC gas_NN1 consumption_NN1 from_II one_MC1 spot_NN1 ._. 
When_CS a_AT1 cell_NN1 phone_NN1 or_CC credit-card_NN1 bill_NN1 arrives_VVZ ,_, each_DD1 call_NN1 or_CC purchase_NN1 is_VBZ itemized_VVN ,_, making_VVG it_PPH1 possible_JJ to_TO track_VVI trends_NN2 in_II calling_VVG or_CC spending_NN1 ,_, which_DDQ is_VBZ especially_RR helpful_JJ if_CS you_PPY use_VV0 a_AT1 phone_NN1 plan_NN1 with_IW limited_JJ minutes_NNT2 or_CC are_VBR trying_VVG to_TO stick_VVI to_II a_AT1 budget_NN1 ._. 
Within_II the_AT next_MD few_DA2 years_NNT2 ,_, household_NN1 utilities_NN2 could_VM be_VBI itemized_VVN as_RR21 well_RR22 ,_, allowing_VVG residents_NN2 to_TO track_VVI their_APPGE usage_NN1 and_CC see_VVI which_DDQ devices_NN2 utilize_VV0 the_AT most_DAT electricity_NN1 ,_, water_NN1 ,_, or_CC gas_NN1 ._. 
New_JJ sensor_NN1 technology_NN1 that_CST consists_VVZ of_IO a_AT1 single_JJ device_NN1 for_IF each_DD1 utility_NN1 ,_, which_DDQ builds_VVZ a_AT1 picture_NN1 of_IO household_NN1 activity_NN1 by_II tracing_VVG electrical_JJ wiring_NN1 ,_, plumbing_NN1 ,_, and_CC gas_NN1 lines_NN2 back_RP to_II specific_JJ devices_NN2 or_CC fixtures_NN2 ,_, could_VM make_VVI this_RG far_RR simpler_JJR to_TO implement_VVI ._. 
Shwetak_VV0 Patel_NP1 ,_, a_AT1 professor_NN1 of_IO computer_NN1 science_NN1 and_CC electrical_JJ engineering_NN1 at_II the_AT University_NN1 of_IO Washington_NP1 ,_, in_II Seattle_NP1 ,_, developed_VVD the_AT sensors_NN2 ,_, which_DDQ plug_VV0 directly_RR into_II existing_JJ infrastructure_NN1 in_II buildings_NN2 ,_, thereby_RR eliminating_VVG the_AT need_NN1 for_IF an_AT1 elaborate_JJ set_NN1 of_IO networked_JJ sensors_NN2 throughout_II a_AT1 structure_NN1 ._. 
For_REX21 example_REX22 ,_, an_AT1 electrical_JJ sensor_NN1 plugs_VVZ into_II a_AT1 single_JJ outlet_NN1 and_CC monitors_VVZ characteristic_JJ "_" noise_NN1 "_" in_II electrical_JJ lines_NN2 that_CST are_VBR linked_VVN to_II specific_JJ devices_NN2 ,_, such_II21 as_II22 cell-phone_JJ chargers_NN2 ,_, refrigerators_NN2 ,_, DVD_NP1 players_NN2 ,_, and_CC light_NN1 switches_NN2 ._. 
And_CC a_AT1 gas_NN1 sensor_NN1 attaches_VVZ to_II a_AT1 gas_NN1 line_NN1 and_CC monitors_VVZ pressure_NN1 changes_NN2 that_CST can_VM be_VBI correlated_VVN to_II turning_VVG on_II a_AT1 stove_NN1 or_CC furnace_NN1 ,_, for_REX21 instance_REX22 ._. 
Now_RT ,_, Patel_NP1 and_CC his_APPGE colleagues_NN2 have_VH0 developed_VVN a_AT1 pressure_NN1 sensor_NN1 that_CST fits_VVZ around_RP a_AT1 water_NN1 pipe_NN1 ._. 
The_AT technology_NN1 ,_, called_VVN Hydrosense_NP1 ,_, can_VM detect_VVI leaks_NN2 and_CC trace_VVI them_PPHO2 back_RP to_II their_APPGE source_NN1 ,_, and_CC can_VM recognize_VVI characteristic_JJ pressure_NN1 changes_NN2 that_CST indicate_VV0 that_CST a_AT1 specific_JJ fixture_NN1 or_CC appliance_NN1 is_VBZ in_II use_NN1 ._. 
Patel_NP1 hopes_VVZ to_TO incorporate_VVI electrical_JJ ,_, gas_NN1 ,_, and_CC water_NN1 sensors_NN2 into_II a_AT1 unified_JJ technology_NN1 and_CC has_VHZ cofounded_VVN a_AT1 soon-to-be-named_JJ startup_NN1 that_CST he_PPHS1 hopes_VVZ will_VM start_VVI offering_VVG combined_JJ smart_JJ meters_NN2 to_II utility_NN1 companies_NN2 within_II the_AT next_MD year_NNT1 or_CC so_RR ._. 
The_AT goal_NN1 ,_, says_VVZ Patel_NP1 ,_, is_VBZ to_TO make_VVI a_AT1 "_" smart_JJ home_NN1 "_" universally_RR deployable_JJ ._. 
"_" I_PPIS1 looked_VVD at_II the_AT existing_JJ infrastructures_NN2 ,_, "_" he_PPHS1 says_VVZ ,_, "_" and_CC saw_VVD that_CST they_PPHS2 could_VM be_VBI retrofitted_VVN ._. "_" 
Smart_JJ sensors_NN2 have_VH0 become_VVN increasingly_RR popular_JJ over_II the_AT past_JJ few_DA2 years_NNT2 as_CSA more_DAR people_NN have_VH0 become_VVN interested_JJ in_II cutting_VVG their_APPGE utility_NN1 bills_NN2 and_CC minimizing_VVG the_AT resources_NN2 that_CST they_PPHS2 consume_VV0 ._. 
A_AT1 number_NN1 of_IO startups_NN2 offer_VV0 to_TO connect_VVI utility_NN1 providers_NN2 and_CC consumers_NN2 so_RR that_DD1 resource_NN1 use_NN1 can_VM be_VBI tracked_VVN over_II the_AT Internet_NN1 ._. 
So_RG far_RR ,_, however_RR ,_, no_AT company_NN1 or_CC utility_NN1 has_VHZ been_VBN able_JK to_TO provide_VVI the_AT sort_NN1 of_IO fine-grain_JJ resource_NN1 usage_NN1 that_CST Patel_NP1 hopes_VVZ to_TO offer_VVI with_IW his_APPGE startup_NN1 ._. 
The_AT idea_NN1 behind_II the_AT water_NN1 sensor_NN1 has_VHZ its_APPGE origins_NN2 in_II Patel_NP1 's_GE original_JJ work_NN1 with_IW electrical_JJ lines_NN2 ._. 
Rather_CS21 than_CS22 simply_RR looking_VVG at_II the_AT amount_NN1 of_IO power_NN1 consumed_VVN by_II all_DB the_AT devices_NN2 in_II a_AT1 house_NN1 ,_, he_PPHS1 decided_VVD to_TO look_VVI at_II noise_NN1 patterns_NN2 --_JJ irregularities_NN2 in_II the_AT electrical_JJ signal_NN1 --_NN1 that_CST propagated_VVD over_II household_NN1 power_NN1 lines_NN2 as_II a_AT1 result_NN1 of_IO electrical_JJ consumption_NN1 ._. 
"_" Let_VM21 's_VM22 say_VVI you_PPY turn_VV0 on_II a_AT1 light_NN1 switch_NN1 in_II the_AT bathroom_NN1 and_CC kitchen_NN1 ,_, "_" he_PPHS1 says_VVZ ._. 
"_" We_PPIS2 can_VM tell_VVI the_AT difference_NN1 between_II the_AT two_MC "_" due_II21 to_II22 electrical_JJ impulses_NN2 that_CST resonate_VV0 at_II a_AT1 high_JJ frequency_NN1 ._. 
"_" So_RR if_CS you_PPY have_VH0 two_MC different_JJ impulses_NN2 you_PPY see_VV0 originate_VV0 from_II two_MC different_JJ locations_NN2 inside_II the_AT home_NN1 ,_, you_PPY can_VM trace_VVI them_PPHO2 back_RP to_II a_AT1 particular_JJ device_NN1 ,_, "_" Patel_NP1 says_VVZ ,_, noting_VVG that_DD1 location_NN1 can_VM be_VBI determined_VVN by_II the_AT amount_NN1 of_IO time_NNT1 that_CST it_PPH1 takes_VVZ for_IF a_AT1 signal_NN1 to_TO reach_VVI the_AT sensor_NN1 ,_, which_DDQ is_VBZ usually_RR just_RR plugged_VVN into_II a_AT1 spare_JJ wall_NN1 outlet_NN1 ._. 
Likewise_RR ,_, Hydrosense_NP1 consists_VVZ of_IO a_AT1 single_JJ device_NN1 attached_VVN to_II a_AT1 cutoff_NN1 valve_NN1 or_CC water_NN1 bib_VV0 that_CST monitors_VVZ the_AT entire_JJ plumbing_NN1 infrastructure_NN1 ._. 
"_" When_CS you_PPY open_VV0 a_AT1 valve_NN1 ,_, the_AT pressure_NN1 on_II the_AT entire_JJ system_NN1 goes_VVZ down_RP ,_, "_" says_VVZ Patel_NP1 ._. 
"_" And_CC whenever_RRQV you_PPY change_VV0 the_AT water_NN1 flow_NN1 from_II static_JJ to_II kinetic_JJ ,_, you_PPY get_VV0 a_AT1 shock_NN1 wave_NN1 that_CST propagates_VVZ throughout_II the_AT pipes_NN2 ._. "_" 
He_PPHS1 explains_VVZ that_CST the_AT shock_NN1 wave_NN1 ,_, while_CS relatively_RR mild_JJ ,_, has_VHZ a_AT1 characteristic_JJ shape_NN1 that_CST can_VM be_VBI used_VVN to_TO identify_VVI different_JJ fixtures_NN2 --_NN1 even_RR the_AT distinction_NN1 between_II the_AT toilets_NN2 in_II different_JJ bathrooms_NN2 ._. 
Using_VVG data_NN collected_VVN in_II nine_MC homes_NN2 of_IO varying_JJ style_NN1 and_CC age_NN1 and_CC with_IW a_AT1 diversity_NN1 of_IO plumbing_NN1 systems_NN2 located_VVN in_II three_MC different_JJ cities_NN2 ,_, Patel_NP1 and_CC his_APPGE colleagues_NN2 have_VH0 shown_VVN that_CST by_II monitoring_VVG these_DD2 shock_NN1 waves_NN2 ,_, it_PPH1 is_VBZ possible_JJ to_TO identify_VVI individual_JJ fixtures_NN2 with_IW 95.6_MC percent_NNU accuracy_NN1 ._. 
"_" The_AT idea_NN1 of_IO being_VBG able_JK to_TO plug_VVI one_MC1 device_NN1 into_II a_AT1 home_NN1 and_CC build_VV0 a_AT1 picture_NN1 of_IO what_DDQ 's_VBZ going_VVG on_RP and_CC off_RP is_VBZ really_RR fascinating_JJ ,_, "_" says_VVZ Adrien_NP1 Tuck_NN1 ,_, CEO_NN1 of_IO Tendril_NP1 Networks_NN2 ,_, a_AT1 company_NN1 that_CST makes_VVZ smart_JJ meters_NN2 and_CC plugs_NN2 for_IF homes_NN2 ._. 
But_CCB he_PPHS1 suspects_VVZ that_CST there_EX will_VM be_VBI some_DD kinks_NN2 to_TO iron_VVI out_RP before_II the_AT technology_NN1 is_VBZ deployable_JJ at_II a_AT1 large_JJ scale_NN1 ._. 
"_" If_CS it_PPH1 were_VBDR easy_JJ ,_, it_PPH1 would_VM have_VHI been_VBN done_VDN already_RR ,_, "_" he_PPHS1 says_VVZ ,_, "_" and_CC that_CST probably_RR means_VVZ that_CST there_EX are_VBR some_DD things_NN2 that_CST need_VV0 to_TO be_VBI teased_VVN out_RP ._. "_" 
In_II31 addition_II32 to_II33 monitoring_VVG utility_NN1 usage_NN1 ,_, Patel_NP1 says_VVZ that_CST the_AT sensors_NN2 can_VM track_VVI human_JJ activity_NN1 within_II a_AT1 home_NN1 ,_, which_DDQ could_VM be_VBI useful_JJ for_IF elder_JJR care_NN1 and_CC reducing_VVG energy_NN1 waste_NN1 ._. 
He_PPHS1 has_VHZ also_RR developed_VVN a_AT1 fourth_MD sensor_NN1 that_CST can_VM be_VBI integrated_VVN into_II a_AT1 home_NN1 's_GE heating_NN1 and_CC cooling_NN1 systems_NN2 ._. 
By_II monitoring_VVG pressure_NN1 changes_NN2 that_CST occur_VV0 when_RRQ people_NN open_VV0 and_CC close_JJ doors_NN2 and_CC when_CS they_PPHS2 enter_VV0 and_CC exit_VV0 a_AT1 room_NN1 ,_, a_AT1 sensor_NN1 within_II an_AT1 air-conditioning_JJ unit_NN1 can_VM infer_VVI with_IW relative_JJ accuracy_NN1 where_RRQ people_NN are_VBR within_II a_AT1 home_NN1 or_CC apartment_NN1 ,_, Patel_NP1 says_VVZ ._. 
Home_NN1 Sensor_NN1 Startup_NN1 Snapped_VVD Up_RP Belkin_NP1 buys_VVZ Zensi_NP1 ,_, which_DDQ makes_VVZ sensors_NN2 to_TO track_VVI domestic_JJ energy_NN1 and_CC water_NN1 usage_NN1 ._. 
If_CS you_PPY knew_VVD how_RGQ much_DA1 electricity_NN1 your_APPGE plasma_NN1 television_NN1 used_VVN or_CC how_RGQ much_DA1 water_NN1 your_APPGE dishwasher_NN1 drank_VVD at_II different_JJ times_NNT2 of_IO day_NNT1 ,_, would_VM you_PPY change_VVI your_APPGE habits_NN2 to_TO conserve_VVI more_DAR and_CC spend_VVI less_RRR on_II utilities_NN2 ?_? 
Researchers_NN2 at_II the_AT University_NN1 of_IO Washington_NP1 ,_, Duke_NN1 University_NN1 ,_, and_CC Georgia_NP1 Tech_NN1 believe_VV0 that_CST you_PPY might_VM ._. 
Several_DA2 years_NNT2 ago_RA they_PPHS2 invented_VVD sensors_NN2 that_CST could_VM track_VVI the_AT electricity_NN1 consumption_NN1 and_CC water_NN1 usage_NN1 throughout_II an_AT1 entire_JJ building_NN1 via_II a_AT1 single_JJ point_NN1 on_II each_DD1 system_NN1 ._. 
In_II 2008_MC ,_, the_AT researchers_NN2 founded_VVD a_AT1 company_NN1 called_VVN Zensi_JJ to_TO commercialize_VVI the_AT technology_NN1 ,_, and_CC last_MD week_NNT1 ,_, they_PPHS2 sold_VVD that_DD1 company_NN1 to_II Belkin_NP1 ,_, an_AT1 electronics_NN1 hardware_NN1 manufacturer_NN1 ._. 
A_AT1 line_NN1 of_IO easy-to-install_VV0 sensors_NN2 for_IF homes_NN2 could_VM be_VBI commercially_RR available_JJ within_II the_AT next_MD year_NNT1 ,_, says_VVZ Shwetak_NP1 Patel_NP1 ,_, professor_NN1 of_IO computer_NN1 science_NN1 and_CC engineering_NN1 at_II the_AT University_NN1 of_IO Washington_NP1 ,_, and_CC co-inventor_NN1 of_IO Zensi_NP1 's_GE sensors_NN2 ._. 
Data_NN from_II such_DA sensors_NN2 could_VM lead_VVI to_II itemized_JJ utility_NN1 bills_NN2 --_NN1 and_CC customers_NN2 who_PNQS are_VBR more_RGR aware_JJ of_IO the_AT energy_NN1 sinks_NN2 in_II their_APPGE homes_NN2 ,_, he_PPHS1 says_VVZ ._. 
Data_NN from_II the_AT electricity_NN1 and_CC water_NN1 sensors_NN2 is_VBZ sent_VVN via_II the_AT Internet_NN1 to_II a_AT1 base_NN1 station_NN1 for_IF analysis_NN1 ._. 
The_AT algorithms_NN2 differentiate_VV0 between_II different_JJ devices_NN2 and_CC calculate_VV0 electricity_NN1 and_CC water_NN1 usage_NN1 ._. 
The_AT technology_NN1 will_VM face_VVI competition_NN1 ,_, says_VVZ Harvey_NP1 Michaels_NP2 ,_, an_AT1 energy-efficiency_JJ scientist_NN1 and_CC lecturer_NN1 at_II MIT_NP1 ._. 
Cisco_NP1 and_CC MIT_VV0 researchers_NN2 ,_, not_XX including_II Michaels_NP2 ,_, are_VBR collaborating_VVG on_II a_AT1 chip_NN1 that_CST can_VM be_VBI built_VVN into_II every_AT1 energy-consuming_JJ object_NN1 in_II a_AT1 house_NN1 ._. 
The_AT chip_NN1 will_VM track_VVI energy_NN1 use_NN1 and_CC communicate_VVI with_IW a_AT1 smart_JJ meter_NN1 to_TO automatically_RR conserve_VVI energy_NN1 and_CC lower_JJR bills_NN2 ._. 
Belkin_NP1 ,_, a_AT1 privately_RR held_VVN company_NN1 ,_, has_VHZ not_XX disclosed_VVN the_AT acquisition_NN1 price_NN1 for_IF Zensi_NN1 ._. 
The_AT startup_NN1 's_GE CEO_NN1 ,_, Kevin_NP1 Ashton_NP1 ,_, will_VM serve_VVI as_II the_AT general_JJ manager_NN1 of_IO its_APPGE Belkin_NP1 's_GE Conserve_VV0 business_NN1 unit_NN1 ,_, which_DDQ will_VM manage_VVI the_AT startup_NN1 's_GE intellectual_JJ property_NN1 ._. 
The_AT acquisition_NN1 came_VVD before_II Zensi_NP1 closed_VVD a_AT1 first_MD round_NN1 of_IO funding_NN1 ._. 
Belkin_NP1 is_VBZ still_RR working_VVG out_RP the_AT product_NN1 's_GE details_NN2 ._. 
Patel_NP1 says_VVZ the_AT interface_NN1 might_VM be_VBI on_II a_AT1 panel_NN1 within_II the_AT home_NN1 ,_, available_JJ via_II a_AT1 website_NN1 ,_, or_CC sent_VVD directly_RR to_II a_AT1 person_NN1 's_GE phone_NN1 ._. 
